- Boost - lists.boost.org

Re: [boost] [msm] scalability
by Christophe Henry 14 Dec '09

14 Dec '09

Hi Andy, >Christophe, I'm sorry if my 2 cents won't be relevant since I'm not very >familiar with neither MSM I'm thankful for any help. >The most common reason for compiler slowliness/crashiness that I've seen >came from too deep of a template recursion. Agreed ;-) >I'm not totally sure, but I >think that you're using mpl::vector, which, I think, is a form of a type list. AFAIK, this would be mpl::list. mpl::vector is defined as (for example for vector4): template< typename T0, typename T1, typename T2, typename T3 > struct vector4 { typedef aux::vector_tag<4> tag; typedef vector4 type; typedef T0 item0; typedef T1 item1; typedef T2 item2; typedef T3 item3; ... }; So there is no recursion depth, at least not here. I think the culprit would more likely be the many mpl::fold. Thanks, Christophe

2 1

[msm] eUML guard/action location
by Christophe Henry 14 Dec '09

14 Dec '09

Hi Michael, I just remembered the last reason for the current eUML syntax: CurrentState() + cool_event() = DestState() [Guard()]/(action()) It is that eUML row actions/guards can become quite long (something like adjacent_find_(...) ) and I was having difficulties keeping the rows readable. Concretely, I often had to keep the target state on a line by itself, which was costing a line of my screen. With the current syntax, you can get it easier: CurrentState() + cool_event() = DestState() [VeeeeeryLoooooooooongGuard()] /(LooooooooooooooooongAction()) instead of: CurrentState() + cool_event() [VeeeeeryLoooooooooongGuard()] /(LooooooooooooooooongAction()) = DestState() For the tranquility of my conscience, are you still sure you prefer the more logical syntax? Christophe

7 7

[chrono] Boost.Chrono version 0.2.1 Bug fixes + updated documentation
by vicente.botet 13 Dec '09

13 Dec '09

Hi, I have updated Boost.Chrono version 0.2.1 with a lot of fixes. Bug Fixes * Replace INTMAX_C by BOOST_INTMAX_C until boost/cstdint;hpp ensures INTMAX_C is always defined. * Define BOOST_CHRONO_HAS_CLOCK_MONOTONIC when BOOST_CHRONO_WINDOWS_API * Commenting invalid operator declarations * Take care of Boost min/max recommendations * complete qualification when defining nested typedef duration on clocks to avoid the following compile error: ./boost/chrono/chrono.hpp:945: error: declaration of 'typedef class boost::chrono::duration<long long int, boost::ratio<1ll, 10000000ll> > boost::chrono::system_clock::duration' ./boost/chrono/chrono.hpp:458: error: changes meaning of 'duration' from 'class boost::chrono::duration<long long int, boost::ratio<1ll, 10000000ll> >' * cleanup of simulated... * warning removal on posix/process_clock.cpp * disable VC++ foolishness * Update Jamfiles to manage with dll * removal of conversion warning in test_duration * manage with MSVC reporting a warning instead of an error when there is an integral constant overflow * Use STATIC_ASSERT specific macro to solve the compile failing issue * Qualify with boost::detail boost::chrono::detail to avoid ambiguities with MSVC Documentation: * Updated documentation. The implementation will eventually work with most C++03 conforming compilers. Tests have been run on Windows with * VC++ 9.0 SP1 Cygwin with * GCC 3.4.4 * GCC 4.3.2 MinGW with * GCC 4.4.0 Getting Boost.Chrono You can get the last stable release of Boost.Chrono by downloading chrono.zip from the Boost Vault You can also access the latest (unstable?) state from the Boost Sandbox. Where to install Boost.Chrono? The simple way is to decompress (or checkout from SVN) the file in your BOOST_ROOT directory. Othesewise, if you decompress in a different directory, you will need to comment some lines, and uncomment and change others in the build/Jamfile and test/Jamfile. Sorry for this, but I have not reached yet to write a Jamfile that is able to work in both environements and use the BOOST_ROOT variable. Any help is welcome. Building Boost.Chrono Boost.Chrono is not a header only library. You need to compile it before use. bjam libs/chrono/build Building an executable that uses Boost.Chrono In addition to link with the Boost Chrono library you need also to link with the Boost System library. Testing In order to test you need to do. bjam libs/chrono/test It will be very kind if someone tested this release with other compilers (Intel) and platforms (Mac) as I'm unable to test them. I expect to been able to test on Linux/gcc-4.1.2 soon. Of course any comments and suggestions are welcome. Best regards, _____________________ Vicente Juan Botet Escribá

1 0

[Spirit/Qi] Sequential-or attribute access in semantic action
by Adam Butcher 13 Dec '09

13 Dec '09

Hi, I am having a bit trouble with the semantic action of a sequential-or parser. Specifically my problem is with accessing the attribute of the sequential-or parser. My first observation is that the documentation at http://www.boost.org/doc/libs/1_41_0/libs/spirit/doc/html/spirit/qi/quick_r… defines the attribute type of (a || b) as tuple<A, B>. Unless I've misunderstood, this must be a typo as they should both be optional (it is never valid for both to be omitted but I would expect that client code be able to determine which out of (a), (b) or (a >> b) were provided.) My real problem is that applying a semantic action to [what I think is] an expression returning a sequential-or parser, results in the _1 referring to the attribute of the last parser on the right-hand-side of the sequential-or rather than the attribute of the sequential-or parser itself. In the following snippet (assume 'start' has an attribute of type S), idxspec = '[' >> (start || (':' >> -start)) [ _val = access_index (_r1,lvalue,_1) ] >> ']'; the function 'access_index' is called with an arg3 of type 'optional<S>' rather than the expected 'tuple<optional<S>, optional<optional<S>>>'. Wrapping the sequential-or expression in a 'repeat' directive as repeat(1)[start || (':' >> -start)] solved the problem but obviously gave me an unwanted sequence (albeit with only one element). The attribute type of the above came out to be vector<tuple<optional<S>, optional<optional<S>>>> as expected. To remove the sequence I created a parser directive called 'identity' following the form of 'repeat' which I have attached. My final rule is now idxspec = '[' >> identity[start || (':' >> -start)] [ _val = access_index (_r1,lvalue,_1) ] >> ']'; which is working fine; the _1 delivers an argument of type tuple<optional<S>, optional<optional<S>>> as required. However, I would have expected this behaviour from the first expression. Is this a a proto issue, a spirit issue or have I completely misunderstood something somewhere? It may or may not be relevant but I am using both spirit.classic and spirit 2.1 in the same file. This is historical and will be fixed in time but I guess it could be important. So far I have had no problems with the combination however. Additional info: I am using gcc 4.4.2, stlport 5.1.3 and boost 1.41.0. Regards, Adam

2 6

[chrono] New Features + Bug fixes + Updated documentation
by vicente.botet 13 Dec '09

13 Dec '09

Hi, I'm pleased to annonce the new version of the Boost.Chrono library. Even if the library is not yet ready for review I wanted to share it with you. The library can be downloaded from the vault http://www.boostpro.com/vault/index.php?action=downloadfile&filename=chrono… and the documentation is online on the sandbox http://svn.boost.org/svn/boost/sandbox/chrono/libs/chrono/doc/html. For the moment I'm not able to test it on Windows and Mac platforms. Could you check if there are some regressions on Windows or on Mac platforms? Of course, any comments or suggestions are welcome. If there is a volontary review manager that can start inspecting the library s/he is also welcome :). This library is based on the Beman's boostified version of the probe of concept of Howard. Many thanks to Howard the originator of the library and to Beman to start the boostification. Next follows the contents of the library and the updates of the new version Version 0.1.0, April 29, 2009 Beman's boostified version Features: * The C++0x Standard Library's class common_type. * The C++0x Standard Library's compile-time rational arithmetic. * The C++0x Standard Library's time utilities, including: * Class template duration * Class template time_point * Clocks: * system_clock * monotonic_clock * high_resolution_clock * Class template timer, with typedefs: * system_timer * monotonic_timer * high_resolution_timer * Process clocks and timers: * process_clock, capturing real, user-CPU, and system-CPU times. * process_timer, capturing elapsed real, user-CPU, and system-CPU times. * run_timer, convenient reporting of process_timer results. Version 0.2.0, December 8, 2009 + New Features + Bug fixes + Updated documentation Features: * Added ratio construction and assignment from an equivalent ratio (LWG 1281. CopyConstruction and Assignment between ratios having the same normalized form) * Added nested ratio typedef type (LWG 1281. CopyConstruction and Assignment between ratios having the same normalized form) * Added BOOST_CHRONO_HAS_CLOCK_MONOTONIC macro to state if monotonic_clock is provided on a platform. * Added duration operator% (LWG 934. duration is missing operator%) * Complete duration operator* and operator/ Implementation: * Use INTMAC_C to name intmax_t constants instead of LL. * Separate chrono.cpp on # files win/chrono.cpp, mac/chrono.cpp and posix/chrono.cpp to make easier the maintenance on different platforms. * Separate process_clock.cpp on # files win/process_clock.cpp, mac/process_clock.cpp and posix/process_clock.cpp to make easir the maintenace on different platforms. * Added the error_code prototype for monotonic_clock::now for mac/chono.cpp * Fully implement mac/chrono.cpp with error handling. * Take care on POSIX systems when CLOCK_MONOTONIC is not defined. Documentation: * The documentation is written now using quick-book using as base "N2661 - A Foundation to Sleep On". Fixed Bugs * operator/ was ambiguous: Disambiguate duration operator/ * CLOCK_MONOTONIC is not defined with cygwin/gcc 3.4: Disable code when BOOST_CHRONO_HAS_CLOCK_MONOTONIC is not defined. * result of metafunctions ratio_multiply and ratio_divide were not normalized ratios: Use of the nested ratio typedef type on ratio arithmetic operations. * Copy constructor from similar duration masked the defaulted operations: Added duration defaulted implementations Note that the comon_type class should be integrated by Boost.TypeTraits, class ratio will be proposed as a standalone library Boost.Ratio, and the Timer and Process clock part could be proposed as an update of Boost.Timer. Best regards, _____________________ Vicente Juan Botet Escribá

4 11

Re: [boost] [msm] Review
by Christophe Henry 13 Dec '09

13 Dec '09

Hi Darryl, If you don't mind, I'll start by answering the last question, it'll make the rest of the discussion easier. >Now I'm confused. Are you: > >a) Convinced the suggestion is without merit? No. After reading your last mail, I see a bit better, though I'm not sure I can follow all of it. >b) Waiting for a concrete example? Yes! Exactly. >c) Convinced I'm an idiot that can't figure out how to use msm effectively >(in which case, consider this a request for better docs)? No. Sorry if I gave you the feeling the correct answer was a) or b). I don't mind criticism, any of it, justified or not, as long as it doesn't stay vague and instead goes to concrete, argumented points. Vague negative comments bring little to the review and contribute to giving an incomfortable feeling to all participants. Constructive remarks are, on the other side, highly welcome! That's reason #1 for a review. I sometimes, like yesterday, take part in several parallel discussions, so I have little time left for remarks I cannot start anything with. Sorry about it, it's the negative side of the review. I'd be thankful if potentially useful but vague ideas were kept for a more quiet time. So now, to the concrete parts of your remarks. >I genuinely don't see why msm needs to instantiate state objects it owns at all. If this is the point, then it is a solvable problem. Clearly, it doesn't strictly need. It's just a feature, which is in some contexts very useful. The main one is to be able to save some data in the states. This data can then be generically accessed by the generic functors, thus increasing reuse and encouraging clean code. The functor front-end will be your friend in this case. The same with eUML, but it doesn't have to be eUML. The second reason would be entry/exit: >I suggest spelling them enter<state>(machine) and exit<state>(machine) > or machine.enter<state>() and machine.exit<state>(). Of course, this would be a solution. It's actually an interesting one, and if you don't mind, we'll keep for later (see below). Do we agree that it makes the machine slightly less readable? The 3rd reason is state reuse. This brings us to: >What does reuse of a state in msm mean? Interesting question, which is sadly ignored by many fsm frameworks. A case would be when you want to avoid doing the same case again and again. You could then define a state hierarchy, even (gulps) OO-like for this (or better some template policy magic). For example, I personally often add some logging in debug mode, so that I know what happened in my system. The visitor is another good example of it. I hope that after having a look at it, you'll see the interest you can gain from it. You might have easily missed another case in the doc. If you have a look at iPodSearch and iPodSearchEuml, you'll see where I want to go next. To states which implement STL algorithms. You will then be able to write complex algorithms by simply moving some predefined states on a diagram. I plan to call it the STL (State Template Library ;-) ). While this is for the future, I'll need independent, reusable states for it. >No. For the record, in my opinion, the single the worst feature of >statechart is the necessity to use custom reactions to implement some fsm <constructs. This was a subject of much discussion during the Statechart >(FSM) review iirc. I see. Then I probably misunderstood your point. > Can you perhaps explain why flags are needed? Sure. If you look at the statechart doc, you'll see some state_cast. I personally find it not beautiful in my code. It's also hard to maintain when your fsm structure changes. Flags are faster, require no rtti, are more model-oriented as states define themselves their "property" and makes user code easier/smaller to read. It's even supported by eUML for better model-orientation. I wanted to handle the entry/exit discussion separately. Hopefully I managed to convince you why states are useful and their instanciation brings some advantages. Of course, there is a price to pay, though a small one (if without data, 1 byte/state). This is in most cases not a huge cost. I agree there are cases where the price is still too expensive. I also agree, that the one current MSM back-end is not a solution for all problems, otherwise I would have written in stone that there is one backend and we'd be done. The MSM backend is a fully optimized machine, trying to pack as many features as possible at the lowest possible cost. There are however cases, where one would be ready to give up some features like state reuse and generic functors in exchange for reduced costs. For such cases, there is room for a mini-MSM backend (name from Barend Gehrels ;-) ). And in this case, your entry/exit solution appears a viable, interesting one for which I am thankful. When preparing the review, I had several possibilities: - make the current MSM backend, with lots of features, thus proving that the approach of the MPL book was a great one. - make a reduced, smaller cost version, with the certainty of being criticized for not having enough features (see review from FSM) and rending a disservice to the book. - make a 3rd one, more dynamic, which I want to do since a while. - make no review until all possible back-ends are done and make everybody wait at least one more year. Do we agree that solution 1 seems the more logical? I hope we managed to find a common ground and that a fruitful discussion can take place. Christophe

2 1

[test] new feature - colored output
by Gennadiy Rozental 13 Dec '09

13 Dec '09

Hi, I've implemented a change in regular human readable log and report outputs. You now have an ability to tell the unit test framework to produce colored output. New boolean runtime parameter is added: Environment variable: BOOST_TEST_COLOR_OUTPUT Command-line long: --color_output Command-line short: -x. By default the feature is disabled. Any comments are appreciated. Specifically: how does this feature works in different environments and if there is a way to make MS compiler to work properly (gcc on windows works fine in Windows shell, while msvc build does not) Gennadiy

1 0

[fiber] new version 0.3.0 with future and fiber-migration
by Oliver Kowalke 12 Dec '09

12 Dec '09

Hi, I've uploaded boost.fiber 0.3.0 - it supports future/packaged_task (adapted from Anthony), migrating fiber between schedulers (event running in different threads) and sync. primitives like barrier etc. The spiining sync. objects can be used for snyc. fibers running in different schedulers (running in different threads). Please see examples: ping_pong_mt, migrate_mt, future_mt and barrier_mt regards, Oliver change.log 0.3.0: ------ - support of barrier (more efficient in on scheduler) and spin::barrier (sync .between threads possible) - support of future and packaged_task (example future_mt) - explicit migration of fibers between two schedulers (even if schedulers are in different threads - see example migrate_mt) - locks replaced by locks implementation from boost.thread - introduction of spin::mutex, spin::condition, spin::auto_reset_event, ... (can be used to sync fibers between threads see ping_pong_mt example) - introduction of mutex, condition, auto_reset_event, ... (objects bounded to one scheduler, but more efficient than spin counter-parts) - refactoring of internal classes

1 0

ACID transactions/flushing/MySQL InnoDB
by Stefan Strasser 12 Dec '09

12 Dec '09

hey, this is not a boost question, but it is for the development of a boost library, so maybe a database expert on the list could help me out? I've been benchmarking some test cases of my library against MySQL InnoDB, which isn't even apples to oranges anymore, but nevertheless gave some insights. what I find surprising is the performance of MySQL for small transactions. while my library spends 90% of its time waiting for the disk to flush a transaction, InnoDB uses 100% CPU and barely uses any disk. I've reduced the calls to fsync() for small transactions to 1 call per transaction after some sequential writing, so I'm wondering how MySQL accomplishes this? by definition there must be a flush per transaction, unless you loosen the requirements of the D in ACID. Is that what InnoDB is doing? the only way I see to reduce flushes even more is combining some transactions. but after a power failure some of those transactions can be lost even though they haven reported as successfully committed. I have not yet examined their source code yet. the MySQL documentation doesn't say anything about loosening Durability. (although you have to look pretty hard to see that the default isolation level isn't Serializable either.) thanks in advance, Stefan

5 17

[Boost-config] Add pgi-9.0 threading support
by Belcourt, Kenneth 12 Dec '09

12 Dec '09

Hi John, I've attached a patch to enable threading support with pgi-9.0 (I'm trying to resurrect the pgi tester). Does this look okay to commit? -- Noel

2 1