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

Platform independent memory mapped [file] IO
by Steve 22 Nov '11

22 Nov '11

I have some very large (TB scale) files, and I want to map fixed-size (say 1Mb) segments from them into memory, for both reading and writing (in different contexts) making maximum use of OS-level caching. The software I'm writing needs to work under Unix/Linux and Windows... performance is critical. I've discovered boost::iostreams::mapped_file_source and boost::iostreams::mapped_file_sink, which provide most of the facilities I'm looking for. The facilities I'd like, but haven't found are: * Forcing a synchronisation of written data to disk ( msync(2) on Unix; FlushViewOfFile on Windows) * Locking files to prevent two processes attempting to write the same file at the same time (or read a newly created file before written data is flushed..) * Controlling attributes of the file at creation time (Unix) Can I do these things using boost::iostreams::mapped_file_sink and boost::iostreams::mapped_file_source? I'm aware of boost::interprocess::basic_managed_mapped_file et al, which suppors a flush() method - but I don't want to manage the files as if they were file-backed heaps... with all the space allocation that entails. I am only interested in efficiently reading/writing to fixed-size blocks of very large files (of a size known from the offset.) Am I overlooking a straightforward strategy, using the existing boost::iostreams library, to establish file-locks and to ensure only consistently written data is read by other processes? Is it appropriate to want to to force synchronisation from code as soon as all the data in a block has been written (avoiding a large latency when, say, 1GB of pages have been updated and the process needs to exit?)

1 0

Why wasn't filesystem v2 removed in boost 1.48?
by Thomas Dziedzic 22 Nov '11

22 Nov '11

hi, I was wondering about the removal of filesystem v2 as stated on: http://www.boost.org/doc/libs/1_48_0/libs/filesystem/v3/doc/index.htm which says that filesystem v2 wont be included in boost 1.48 and later. but it seems that boost 1.48.0 still has filesystem v2 and the changelog has nothing about filesystem v2 for boost 1.48 did release guys for boost 1.48 forget to remove filesystem v2? If it was decided to keep boost filesystem v2 for 1.48, could the documentation be changed to reflect that? Thanks for your time,

2 1

Re: [Boost-users] Boost.Locale build missing usage-requirements
by Artyom Beilis 22 Nov '11

22 Nov '11

--- On Fri, 11/18/11, Alexey Badyaev <leemouse(a)mail.ru> wrote: > From: Alexey Badyaev <leemouse(a)mail.ru> > > and user of > library must define symbol BOOST_LOCALE_DYN_LINK=1 in own > project.... Yes... Unless you integrate the full Boost tree into your project and use it directly. Which I don't see any specific reason to do. I'll review and consider adding if this **is** common and practice for boost libraries. > I wrote email to author of library (Artyom Beilis), but he > reject my > proposal of add usage-requirements to lib target. I had seen your note about useage-requirements besides and I have not responded you due the lack of time. And now you write (quoting): > he reject my proposal of add usage-requirements to lib target. This is not only **incorrect**, it is also disrespectful to behave like this and appeal to Boost community just because you had not received an answer from me withing _1_ (one) day. > May anybody with right-access to boost repository fix > this??? Now... I can tell you one thing. Do you really expect me to help you to resolve your problems after you behave this way? I'll get to check your proposal when I'll have enough time to study the issue (among other things I need to do) and I'll see if it is required. Meanwhile, patch your own code... In any case, as you told me you use Boost-Trunk, so this would not be a problem for you... Artyom Beilis -------------- CppCMS - C++ Web Framework: http://cppcms.sf.net/ CppDB - C++ SQL Connectivity: http://cppcms.sf.net/sql/cppdb/

2 1

Boost.Locale build missing usage-requirements
by Alexey Badyaev 22 Nov '11

22 Nov '11

Hi All. Library "Locale" from "Boost C++ Libraries" package has Jamfile.v2 in folder "build". Target "boost_locale" of library in that jam-file defined see below: lib boost_locale : encoding/codepage.cpp shared/date_time.cpp shared/format.cpp shared/formatting.cpp shared/generator.cpp shared/ids.cpp shared/localization_backend.cpp shared/message.cpp shared/mo_lambda.cpp util/codecvt_converter.cpp util/default_locale.cpp util/info.cpp util/locale_data.cpp : # Don't link explicitly, not required <define>BOOST_THREAD_NO_LIB=1 <link>shared:<define>BOOST_LOCALE_DYN_LINK=1 <threading>multi # Meanwhile remove this <conditional>@configure ; You can see: usage-requirements of target is not defined, and user of library must define symbol BOOST_LOCALE_DYN_LINK=1 in own project.... I wrote email to author of library (Artyom Beilis), but he reject my proposal of add usage-requirements to lib target. May anybody with right-access to boost repository fix this??? -- With best regards, Alexey Badyaev Russian Federation, Kemerovo

2 2

boost MPI: unsupported functions & support on super computers ?
by Avi Bahra 21 Nov '11

21 Nov '11

I am learning MPI, and have a few questions whether its worth using boost MPI, since I need to run on many flavours of unix.(linux clusters and AIX as a minimum) o In the boost MPI documentation it appears that many MPI functions are not supported (MPI_Bsend,etc). If I need this capability then does this mean I should not use boost MPI ? (or is there still a back door way of calling with the C API instead ? ) o Has any one managed to use boost MPI on Super-computers like: - IBM/AIX. - Fujitsu - NEC Since we need to be able swap vendors. Any help appreciated. Ta, Avi

3 2

[unordered] gcc complains in unordered/detail/table.hpp
by Ioannis Papadopoulos 21 Nov '11

21 Nov '11

Hi, In unordered/detail/table.hpp:177 the following exists: iterator& operator++() { node_ = node_ = static_cast<node_pointer>(node_->next_); return *this; } iterator operator++(int) { iterator tmp(node_); node_ = node_ = static_cast<node_pointer>(node_->next_); return tmp; } gcc 4.6.1 complains about boost_1_48_0/include/boost/unordered/detail/table.hpp:177:13: warning: operation on ‘this->boost::unordered::iterator_detail::iterator<boost::unordered::detail::ptr_node<std::pair<const int, int> >*, std::pair<const int, int> >::node_’ may be undefined [-Wsequence-point] what is the purpose of node_ = node_ ?

2 1

[gil] metafunction for layout
by Sapp, Mark 21 Nov '11

21 Nov '11

Hi All, The following code is straight out of the tutorial about making function generic. Can anyone suggest how I might determine the layout of DstView here so that I can build an image of the same layout? In this example they hardcode gray_layout_t and I'm not seeing any metafunctions that return the layout. template <typename SrcView, typename DstView> void x_luminosity_gradient(const SrcView& src, const DstView& dst) { typedef typename channel_type<DstView>::type d_channel_t; typedef typename channel_convert_to_unsigned<d_channel_t>::type channel_t; typedef typename image_type<channel_t, gray_layout_t>::type gray_image_t; typedef typename gray_image_t::value_type gray_pixel_t; gray_image_t ccv_image(src.dimensions()); copy_and_convert_pixels(src, view(ccv_image)); x_gradient(const_view(ccv_image), dst); } Thanks, Mark Notice: This e-mail is intended solely for use of the individual or entity to which it is addressed and may contain information that is proprietary, privileged, company confidential and/or exempt from disclosure under applicable law. If the reader is not the intended recipient or agent responsible for delivering the message to the intended recipient, you are hereby notified that any dissemination, distribution or copying of this communication is strictly prohibited. If this communication has been transmitted from a U.S. location it may also contain data subject to the International Traffic in Arms Regulations or U.S. Export Administration Regulations and cannot be disseminated, distributed or copied to foreign nationals, residing in the U.S. or abroad, without the prior approval of the U.S. Department of State or appropriate export licensing authority. If you have received this communication in error, please notify the sender by reply e-mail or collect telephone call and delete or destroy all copies of this e-mail message, any physical copies made of this e-mail message and/or any file attachment(s).

2 4

Problems linking shared library with ASIO/System/Serialization/Python
by Federico J. Fernández 21 Nov '11

21 Nov '11

Hi All, I'm trying to build a shared library on Linux, that will become a Python module via Boost.Python. That library also requires functionality from Boost.ASIO and Boost.Serialization. I'm using Boost 1.46.1, and gcc version 4.5.2 (Ubuntu/Linaro 4.5.2-8ubuntu4). I'm getting undefined errors related to Boost.System during the compilation: g++ -L/usr/local/lib/ -shared python_client.o network_interface.o geometry_model_api.o client.o -o python_client.so -lpython2.7 -lboost_python -lboost_serialization -lboost_thread -lQtCore -lpthread -Wl,-no-undefined network_interface.o: In function `get_system_category': /usr/local/include/boost/asio/error.hpp:220: undefined reference to `boost::system::system_category()' /usr/local/include/boost/asio/error.hpp:220: undefined reference to `boost::system::system_category()' /usr/local/include/boost/asio/error.hpp:220: undefined reference to `boost::system::system_category()' /usr/local/include/boost/asio/error.hpp:220: undefined reference to `boost::system::system_category()' network_interface.o: In function `error_code': /usr/local/include/boost/system/error_code.hpp:315: undefined reference to `boost::system::system_category()' network_interface.o:/usr/local/include/boost/asio/error.hpp:220: more undefined references to `boost::system::system_category()' follow client.o: In function `__static_initialization_and_destruction_0': /usr/local/include/boost/system/error_code.hpp:214: undefined reference to `boost::system::generic_category()' /usr/local/include/boost/system/error_code.hpp:215: undefined reference to `boost::system::generic_category()' /usr/local/include/boost/system/error_code.hpp:216: undefined reference to `boost::system::system_category()' client.o: In function `get_system_category': /usr/local/include/boost/asio/error.hpp:220: undefined reference to `boost::system::system_category()' client.o: In function `error_code': /usr/local/include/boost/system/error_code.hpp:315: undefined reference to `boost::system::system_category()' /usr/local/include/boost/system/error_code.hpp:315: undefined reference to `boost::system::system_category()' /usr/local/include/boost/system/error_code.hpp:315: undefined reference to `boost::system::system_category()' client.o:/usr/local/include/boost/system/error_code.hpp:315: more undefined references to `boost::system::system_category()' follow collect2: ld returned 1 exit status I tried defining BOOST_SYSTEM_NO_DEPRECATED, some errors disappear, others don't. Any advice is greatly appreciated. Thanks in advance. -- federico

2 2

Trouble building boost-1.48.0 with mingw 4.5.2
by Jose Carlos Rodríguez Rodríguez 21 Nov '11

21 Nov '11

Hi! My first e-mail to the list. :) I am Spanish. Sorry for my English, :/ Boost is a blessing. I have not words. I like to build the library in Windows 7 (VC2010, MingW with gcc-4.5.2) and Linux (Ubuntu 11.10 with gcc-4.6.1). No problem with VC2010. No problem with Ubuntu 11.10 with gcc-4.6.1. However i am unable with MinGW gcc-4.5.2. My steps (works with boost-1.47.0 perfectly(?))): 1. Open mSYS 2. ./bootstrap.sh --with-toolset=mingw 3. Open project-config.bjam and modify... # Compiler configuration. This definition will be used unless # you already have defined some toolsets in your user-config.jam # file. - if ! mingw in [ feature.values <toolset> ] - { - using mingw ; - } - project : default-build <toolset>mingw ; + project : default-build <toolset>gcc ; 3. mkdir stage 4. bjam -a --build-dir="./stage/" --build-type=complete toolset=gcc stage I get: - has_icu builds : no warning: Graph library does not contain MPI-based parallel components. note: to enable them, add "using mpi ;" to your user-config.jam - iconv (libc) : no - iconv (separate) : yes - icu : no - icu (lib64) : no - g++ -shared-* supported : yes - ../config//has_gcc_visibility builds : no - ../config//has_long_double_support builds : yes warning: skipping optional Message Passing Interface (MPI) library. note: to enable MPI support, add "using mpi ;" to user-config.jam. note: to suppress this message, pass "--without-mpi" to bjam. note: otherwise, you can safely ignore this message. warning: No python installation configured and autoconfiguration note: failed. See http://www.boost.org/libs/python/doc/building.html note: for configuration instructions or pass --without-python to note: suppress this message and silently skip all Boost.Python targets V:/OFI-SJLJ-GCC452-32b/MinGW/msys/1.0/home/IUCTC/boost_1_48_0/tools/build/v2/build\virtual-target.jam:1079: in virtual-target.register-actual-name from module virtual-target error: Duplicate name of actual target: <pstage\lib>libboost_exception-mgw45-mt-1_48.a error: previous virtual target { common%common.copy-libboost_exception-mgw45-mt-1_48.a.STATIC_LIB { gcc%gcc.archive-libboost_exception-mgw45-mt-1_48.a.STATIC_LIB { gcc%gcc.compile.c++-clone_current_exception_non_intrusive.o.OBJ { clone_current_exception_non_intrusive.cpp.CPP } } } } error: created from ./stage-proper error: another virtual target { common%common.copy-libboost_exception-mgw45-mt-1_48.a.STATIC_LIB { gcc%gcc.archive-libboost_exception-mgw45-mt-1_48.a.STATIC_LIB { gcc%gcc.compile.c++-clone_current_exception_non_intrusive.o.OBJ { clone_current_exception_non_intrusive.cpp.CPP } } } } error: created from ./stage-proper error: added properties: <linkflags>-shared-libgcc <linkflags>-shared-libstdc++ error: removed properties: none V:/OFI-SJLJ-GCC452-32b/MinGW/msys/1.0/home/IUCTC/boost_1_48_0/tools/build/v2/build\virtual-target.jam:490: in actualize-no-scanner from module object(file-target)@4649 V:/OFI-SJLJ-GCC452-32b/MinGW/msys/1.0/home/IUCTC/boost_1_48_0/tools/build/v2/build\virtual-target.jam:135: in object(file-target)(a)4649.actualize from module object(file-target)@4649 V:/OFI-SJLJ-GCC452-32b/MinGW/msys/1.0/home/IUCTC/boost_1_48_0/tools/build/v2\build-system.jam:749: in load from module build-system V:\OFI-SJLJ-GCC452-32b\MinGW\msys\1.0\home\IUCTC\boost_1_48_0\tools\build\v2/kernel\modules.jam:283: in import from module modules V:\OFI-SJLJ-GCC452-32b\MinGW\msys\1.0\home\IUCTC\boost_1_48_0\tools\build\v2\kernel\bootstrap.jam:142: in boost-build from module V:\OFI-SJLJ-GCC452-32b\MinGW\msys\1.0\home\IUCTC\boost_1_48_0\boost-build.jam:17: in module scope from module However, with flag "minimal" 4. bjam -a --build-dir="./stage/" --build-type=minimal toolset=gcc stage Performing configuration checks - has_icu builds : no warning: Graph library does not contain MPI-based parallel components. note: to enable them, add "using mpi ;" to your user-config.jam - iconv (libc) : no - iconv (separate) : yes - icu : no - icu (lib64) : no - ../config//has_gcc_visibility builds : no - ../config//has_long_double_support builds : yes warning: skipping optional Message Passing Interface (MPI) library. note: to enable MPI support, add "using mpi ;" to user-config.jam. note: to suppress this message, pass "--without-mpi" to bjam. note: otherwise, you can safely ignore this message. warning: No python installation configured and autoconfiguration note: failed. See http://www.boost.org/libs/python/doc/building.html note: for configuration instructions or pass --without-python to note: suppress this message and silently skip all Boost.Python targets Component configuration: - chrono : building - date_time : building - exception : building - filesystem : building - graph : building - graph_parallel : building - iostreams : building - locale : building - math : building - mpi : building - program_options : building - python : building - random : building - regex : building - serialization : building - signals : building - system : building - test : building - thread : building - timer : building - wave : building ...patience... ...patience... ...patience... ...patience... ...found 6852 targets... ...updating 913 targets... common.mkdir stage\lib common.mkdir stage\boost\bin.v2\libs\regex\build\gcc-mingw-4.5.2\debug\link-static common.mkdir stage\boost\bin.v2\libs\regex\build\gcc-mingw-4.5.2\debug\link-static\threading-multi gcc.compile.c++ stage\boost\bin.v2\libs\regex\build\gcc-mingw-4.5.2\debug\link-static\threading-multi\c_regex_traits.o gcc.compile.c++ stage\boost\bin.v2\libs\regex\build\gcc-mingw-4.5.2\debug\link-static\threading-multi\cpp_regex_traits.o gcc.compile.c++ stage\boost\bin.v2\libs\regex\build\gcc-mingw-4.5.2\debug\link-static\threading-multi\cregex.o gcc.compile.c++ stage\boost\bin.v2\libs\regex\build\gcc-mingw-4.5.2\debug\link-static\threading-multi\fileiter.o gcc.compile.c++ stage\boost\bin.v2\libs\regex\build\gcc-mingw-4.5.2\debug\link-static\threading-multi\icu.o gcc.compile.c++ stage\boost\bin.v2\libs\regex\build\gcc-mingw-4.5.2\debug\link-static\threading-multi\instances.o gcc.compile.c++ stage\boost\bin.v2\libs\regex\build\gcc-mingw-4.5.2\debug\link-static\threading-multi\posix_api.o ... ... ... works! But...I need a full building. :( Help! Thank you in advance. :) JC

1 0

Re: [Boost-users] some warning C4273: inconsistent dll linkage
by code 21 Nov '11

21 Nov '11

Hi Alex, I still got some of warning C4273: inconsistent dll linkage like the following ---------------------- 1> thread.cpp1>..\Boost\thread.cpp(186): warning C4273: 'boost::thread::thread' : inconsistent dll linkage1> C:\3rdParty\BOOST_LIBRARY\boost-1.47.0\boost/thread/detail/thread.hpp(154) : see previous definition of '{ctor}'1>..\Boost\thread.cpp(189): warning C4273: 'boost::thread::start_thread' : inconsistent dll linkage1> C:\3rdParty\BOOST_LIBRARY\boost-1.47.0\boost/thread/detail/thread.hpp(120) : see previous definition of 'start_thread'1>..\Boost\thread.cpp(200): warning C4273: 'boost::thread::thread' : inconsistent dll linkage1> C:\3rdParty\BOOST_LIBRARY\boost-1.47.0\boost/thread/detail/thread.hpp(122) : see previous definition of '{ctor}'1>..\Boost\thread.cpp(242): warning C4273: 'boost::thread::~thread' : inconsistent dll linkage1> C:\3rdParty\BOOST_LIBRARY\boost-1.47.0\boost/thread/detail/thread.hpp(155) : see previous definition of '{dtor}' -------------- BTW, *after* I built boost thread libraries, I set #define BOOST_THREAD_DYN_LINK in boost-1.47.0/boost/boost/config/user.hpp, should I set BOOST_THREAD_DYN_LINK in this file *BEFORE* building boost thread libraries? Is it correct? Did I do something wrong? Thank you for your great help! Yan --- On Fri, 11/18/11, code <b4code(a)yahoo.com> wrote: From: code <b4code(a)yahoo.com> Subject: RE: [Boost-users] some warning C4273: inconsistent dll linkage To: "Alex Perry" <Alex.Perry(a)smartlogic.com> Date: Friday, November 18, 2011, 1:12 PM Thank you so much, Alex !! I understand better. Yan --- On Fri, 11/18/11, Alex Perry <Alex.Perry(a)smartlogic.com> wrote: From: Alex Perry <Alex.Perry(a)smartlogic.com> Subject: RE: [Boost-users] some warning C4273: inconsistent dll linkage To: "code" <b4code(a)yahoo.com> Date: Friday, November 18, 2011, 9:43 AM Hi Yan, Probably not : - ) It actually depends on your toolset / platform that you are using. #define BOOST_THREAD_NO_LIB like its version for all the boost libraries #BOOST_ALL_NO_LIB switches of the automagical linking of libraries for boost. Personally on platforms that support this (msvc, borland etc) I find it really useful – give your project the libdir of the boost libs and let the headers figure out what the name of the lib it wants to link to is (in fact I liked it so much I ended up nicking the boost naming conventions and using them for all our internal company library code coz it just made life so much easier – no more worrying whether dependent lib was compiled with /MD, /MDd or /MT etc which is such a pain with windows based compilers – especially when using dll’s rather than static libs) The NO_LIB define is there to avoid this automatic naming (ie the #pragma comment(lib,) the header adds to your code if supported by your toolset) and so it lets you name the lib you want to link to manually in your makefile / solution as per non boost libs (or when using gcc … ). For example in some x-platform code if you wanted the makefile the same for linux/windows (using cygwin or some such). Alex From: code [mailto:b4code@yahoo.com] Sent: 18 November 2011 16:07 To: boost-users(a)lists.boost.org; Alex Perry Cc: b4code(a)yahoo.com Subject: Re: [Boost-users] some warning C4273: inconsistent dll linkage Hi Alex, Thank you so much for the great advice. When I set #define BOOST_THREAD_DYN_LINK in my code, do I need to set #define BOOST_THREAD_NO_LIB at the same time? Thank you so much again! Yan --- On Fri, 11/18/11, Alex Perry <Alex.Perry(a)smartlogic.com> wrote: From: Alex Perry <Alex.Perry(a)smartlogic.com> Subject: [Boost-users] some warning C4273: inconsistent dll linkage To: "boost-users(a)lists.boost.org" <boost-users(a)lists.boost.org> Cc: "b4code(a)yahoo.com" <b4code(a)yahoo.com> Date: Friday, November 18, 2011, 1:11 AM On Thu, 17 Nov 2011 12:07:57 -0800 (PST) code <b4code(a)yahoo.com> wrote:- > Hi Boost Users, > > When I compiled and linked my projects with Boost libraries, I got the > following warnings > ...snip... > I want to link Boost.thread dll dynamically and other Boost libraries statically. > > If I set #define BOOST_ALL_DYN_LINK, I could not generate ...snip... I didn't see any answer to this but apologies if its already been answered. #define BOOST_ALL_DYN_LINK Is for use within your own consuming code - use the bjam --link option to control which version of boost libraries are built eg in boost root do:- bjam variant=debug,release threading=multi link=shared stage --with-thread which would build boost-thread as dll (bjam --help to get more info on options) then: #define BOOST_THREAD_DYN_LINK In your code would use the dll version of boost.thread (any other boost libs would still be looking for static versions) HTH Alex

1 0