- Boost - lists.boost.org

[msvc] disable 64-bit warning
by Jürgen Hunold 18 Sep '08

18 Sep '08

Hi ! When compiling Boost with address-model=64, some libraries issue msvc-warning BOOST_ROOT\boost/config/abi_prefix.hpp(19) : warning C4103: 'BOOST_ROOT\boost\config\abi_prefix.hpp' : alignment changed after including header, may be due to missing #pragma pack(pop) See msvc_9_4103.txt for a full dump using bjam -a -q address-mode=64 -WX Those libraries are: - system - filesystem - serialization - wave Please find the necessary patches attached. Additionaly, I've added _SCL_SECURE_NO_WARNINGS to serialization and changed wave to always set _SCL_SECURE_NO_DEPRECATE and _CRT_SECURE_NO_DEPRECATE regardless of the compiler version. This should silence those warning for the msvc incarnation, too. Okay to commit ? Comments ? Yours, Jürgen -- * Dipl.-Math. Jürgen Hunold ! Ingenieurgesellschaft für * voice: ++49 511 262926 57 ! Verkehrs- und Eisenbahnwesen mbH * fax : ++49 511 262926 99 ! Lister Straße 15 * juergen.hunold(a)ivembh.de ! www.ivembh.de * * Geschäftsführer: ! Sitz des Unternehmens: Hannover * Prof. Dr.-Ing. Thomas Siefer ! Amtsgericht Hannover, HRB 56965 * PD Dr.-Ing. Alfons Radtke !

2 3

Unused Parameter Warnings
by Michael Marcin 18 Sep '08

18 Sep '08

I'm trying to silence a lot of unused parameter warnings in the boost 1.36 libraries I'm using. I'd like to submit these as patches so I don't have to fix them again (I've had to do it for 1.34 and 1.35). Is there a preferred boost wide solution to silencing these warnings? Should I consult each library author on their preference? AFAIK there are 4 common ways to silence these warnings. * Compiler specific pragmas * Pass by const reference to an empty inline template function * Cast to void * Delete the parameter name * C-style comment out the parameter name -- Michael Marcin

3 2

boost::filesystem: leaf() missing?
by Robert Dailey 17 Sep '08

17 Sep '08

Hi, I noticed I cannot use the leaf() function off of boost::filesystem::path anymore in 1.36. The documentation examples are still using leaf(). Is this broken?

2 3

[threadpool]thread management strategies
by vicente.botet 17 Sep '08

17 Sep '08

Hi, I have do a comparaison of the three thread management strategies fixed, lazy and adaptive header files and I found some incoherences that are possible bugs. * In fixed.hpp and lazy.hpp the size_ function should be const std::size_t size_() { return worker_.size(); } * In lazy.hpp submit_ is not used channel_iterator submit_( channel_item const& itm) { return channel_.put( itm); } * In adaptive and lazy add the assertion on constructor BOOST_ASSERT( lwm <= hwm); * In adaptive void core_size( std::size_t size) { shared_lock< shared_mutex > lk( mtx_worker_); core_size_ = size; } must use a unique_lock instead of a shared_lock * In lazy std::size_t max_size() { return max_size_; } should't have a shared_lock guard. The three files have more in common than differences. Doesn't the following pools behaves identically boost::tp::pool< boost::tp::fixed, boost::tp::unbounded_channel< boost::tp::fifo > > pool( boost::tp::max_poolsize( 10) ); boost::tp::pool< boost::tp::lazy< depend_on_core >, boost::tp::unbounded_channel< boost::tp::fifo > > pool( boost::tp::preallocate( 10), boost::tp::core_poolsize( 10), boost::tp::max_poolsize( 10) ); boost::tp::pool< boost::tp::adaptive< depend_on_core, keep_core_size >, boost::tp::unbounded_channel< boost::tp::fifo > > pool( boost::tp::preallocate( 10), boost::tp::core_poolsize( 10), boost::tp::max_poolsize( 10), boost::posix_time::very_high ); Looking at the differences we have the following feature model PoolSize: fixed|variable[AdjustmentPolicy, ShrinkPolicy] ShrinkPolicy: no_shrink | shrink [RecreatePolicy] I was wondering if we can define a class thread_management taking a feature expression and use it as follows thread_management<fixed> to mean the current fixed thread_management<variable<AdjustmentPolicy> > to mean the current lazy<AdjustmentPolicy> thread_management<variable<AdjustmentPolicy, shrink<RecreatePolicy> > > The advantage of this approach is that we avoid the incoherences. The liability is that this class must use metaprogramming in its implementation to preserv the efficiency. What do you think? Vicente

2 2

[threadpool] remarques to the documentation
by vicente.botet 17 Sep '08

17 Sep '08

Hi, I'm reading the documentation and I don't find any description of the concepts Strategy, Channel, QueueingPolicy, Callable. Do you plan to describe explicitly these concepts in the documentation ? What do you think about adding a Worker concept, and adding it as parameter to the Strategy template class? I think that the tutorial must be reworked, including more practical examples, and starting from the most basic use and going on each one of the use cases where the different features are better adapted, currently it seams more to a informal reference manual. Some naming suggestions: * ThreadManagementStrategy sould be a better name for Strategy. * As QueueingPolicy can be either fifo or lifo, SchedulingPolicy sould be more apropiated. * Pending tasks can be more adecuated than queued tasks. You can use take instead of dequeue which is more general. Could you give a use case for lifo order? What about a manual thread management strategy, allowing the user to add new workers or interrupting them? A copy/paste: In the fixed example there is a reference to tp::channel which should be tp::bounded_channel. In the lazy example // creates a lazy pool with unbounded channel // tasks are processed in FIFO order // the pool contains ten worker threads // no worker threads are started at construction boost::tp::pool< boost::tp::lazy< depend_on_core >, boost::tp::unbounded_channel< boost::tp::fifo > > pool( boost::tp::core_poolsize( 5) boost::tp::max_poolsize( 10) ); Can this pool have more than 5 worker threads? If the worker threads are created "if current poolsize size is less than core_poolsize or the channel is full", as the channel is unbounded_channel the channel will never be full, so when the current poolsize size will be 5 no new threads will be created, ins't it? the adaptive ThreadManagementStrategy seems to be in contradiction to the goal of the threadpool "Using a thread pool over creating a new thread for each task may result in better performance and better system stability because the overhead for thread creation and destruction is negated." Could you present a use case in which this adaptive strategy could be useful? In channnel section you say "If the channel becomes empty all worker threads are set to sleep until a new task is enqueued." Except for the adaptive strategy which has a keep_alive timeout, the other waits until a new task is submited, isn't it? Which are the advantages/liabilities of a bounded channel respect to an unbounded one? When it is better to use one of them? It will be great to have an example justifying the need for the rendezvous channel. A copy/paste: In the smart section smart must replace priority template< typename Attr, typename Ord, typename Enq, typename Deq > struct priority A typo: replace terminateing by terminating in docs and in code Best regrads, Vicente

2 2

[intrusive] making a list safe for concurrent readers
by Michael Marcin 17 Sep '08

17 Sep '08

I have a program that protects an object containing a linked with what is essentially a shared_mutex. I'd like to take advantage of the logical constness of traversing a linked list with const iterators and allow concurrent traversal protected by the external shared_mutex. Unfortunately the standard makes no such guarantees on std::list so I've determined I'll have to write my own linked list with a slightly stronger guarantee that concurrent read access is safe. I remember reading during the review that intrusive could be used to make high quality non-intrusive data structures easily. Can I use intrusive to make a std::list like container that is safe to read concurrently? Thanks, Michael Marcin

3 3

[ATTN Beman] request permission to merge Proto to release
by Eric Niebler 17 Sep '08

17 Sep '08

I'd like to merge Proto from trunk to release. The API is stable, the tests are now passing on most toolsets (though it's hard to see from the test results at the moment because of a Boost.Test bug), and much of the review feedback about the docs has been accommodated. I still plan to make doc improvements before 1.37, but I don't think that should hold up the show. OK to proceed? -- Eric Niebler BoostPro Computing http://www.boostpro.com

6 11

Re: [boost] Answers re BENUM library for automated enum streaming
by Tony D 16 Sep '08

16 Sep '08

From: David Abrahams <dave(a)boostpro.com> > > [Iony wrote]: > > [BENUM does need variadic macros] So, limited portability - > > unless your preprocessor library has some relevant workarounds? > I don't have a preprocessor library. For C98 compatibility > Boost.Preprocessor would probably support > > BENUM(X, (A = 1)(B)(C = 22)(D = B|N)); Sorry - had it in my head that Boost.Preprocessor was one of yours... apologies to Vesa Karvonen and Paul Mensonides. Anyway, good to know some possibility exists, though half the appeal of BENUM was that the notation (sans markup) resembled normal enums, facilitating uptake and retrofitting, so not sure if it's worth complicating the implementation to have the notation above available on a proportion of compilers. Happy to hear any opinions, and will look into it if I have spare time (my introspective pre-processor project takes priority unless there's significant demand). Regards, Tony

1 0

Re: [boost] C++ Builder 2009 and Boost compliance
by Nicola Musatti 16 Sep '08

16 Sep '08

siliconman wrote: [...] > Yeah, the new version of C++Builder includes a custom patched version > of boost that is tweaked so that spirit and serialization will build. I > saw that Nicola has started the process of moving the custom patches > into the main boost trunk, so this should start to get better results > soon. Yep. As Boost.Config is mostly patched it should be possible to run meaningful regression tests now, even though results are not going to look encouraging until more of the patches get committed. Cheers, Nicola -- Nicola.Musatti <at> gmail <dot> com Home: http://nicola.musatti.googlepages.com/home Blog: http://wthwdik.wordpress.com/

1 0

[move] Moving the move headers into boost/detail
by Daniel James 16 Sep '08

16 Sep '08

Hello all, A cut down version of the move library in sandbox was in 'boost/unordered/detail' in 1.36 and is pretty stable. I'd like to move it into 'boost/detail' so that other libraries can use it, and probably add support for more object types as well. The full library should be put up for review eventually, but until then it's a useful implementation detail. Or should I leave it in unordered for now? There's nothing stopping anyone from including it directly. I'd also like to add more tests, if it's in detail, would it be appropriate to add them to the unordered library's tests? Or somewhere else? Also, what should I do for documentation? There's some written (thanks to its Adobe roots, so anyone who noticed that the unordered review mainly consisted of editting the documentation can breath a sigh of relief) so it'd be a shame to not make it available. Maybe I should put it up outside of boost? Obviously I'll make it clear that since the library is in detail, the interface should not be considered stable. Buyer beware. Daniel

1 0