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

Re: pool_allocator problems on VC7.1 (real rele ase)
by scleary＠jerviswebb.com 26 Apr '03

26 Apr '03

> -----Original Message----- > From: John Harris [mailto:john.harris@tradingtechnologies.com] > > I can test this for you tomorrow (I'm about to go drinking now). In > return, could you possibly tell me how to change pool_allocator so > that it uses my own tag? My goal is to have a totally local > allocator for some special parts of the code. Why, BTW, does it have > to use the Singleton pool interface instead of the object pool > interface? The latter, I think is what I'm after. I'm going to start with a brief overview of the Boost.Pool interfaces: pool<> (pool.hpp): "The" pool interface. Provides allocation and deallocation of memory. object_pool<> (object_pool.hpp): Restricts access to the pool<> interface. Provides allocation and deallocation of memory for a single object type; and, when destroyed, will call the destructor for any non-deallocated objects allocated from that object_pool. singleton_pool<> (singleton_pool.hpp): Provides access to the pool<> interface as a singleton, distinguished mainly by size of the memory allocations and a tag. pool_allocator<> (pool_alloc.hpp): Restricts access to the singleton_pool<> interface. Provides a Standard-compliant Allocator interface. To answer the specific questions: 1) The pool_allocator<> (and fast_pool_allocator<>) can be relatively easily extended to use other tags for the singleton_pool<> interface underlying them - pass an additional template parameter (after the UserAllocator parameter), defaulting to the currently-used tag. This is a good idea, and I'll put it in when I can. 2) pool_allocator<> uses the singleton_pool<> interface instead of the object_pool<> interface for the following reasons: A) Standard Allocators only handle the allocation and deallocation of memory, and do not deal with actual object types [+]. The intent of object_pool<> is to cleanly call destructors for objects that were never deallocated. Such an action: 1) is unneeded (since all containers should free whatever memory they allocate, and destroy any objects they construct), and 2) would cause the Allocator to get into construction/destruction semantics, which is contrary to the Standard Allocator concept. B) Standard-compliant Allocators that can be used with STL-provided containers must be stateless. This requires an underlying singleton interface. Encapsulating a non-singleton interface within the Allocator, or even sharing one with the handle/body idiom does not satisfy the stateless requirement. [+] The object-related Allocator requirements (construct()/destroy()) do not have anything to do with allocation/deallocation, and are generally considered to be a useless overspecification. Please let me know if I haven't been clear enough - I've been a little lacking in sleep lately ;) -Steve

2 1

RE: [Boost-Users] NRVO (was Re: "Correct" way to return empty shared_ptr?)
by dick.bridges＠tais.com 25 Apr '03

25 Apr '03

return shared_ptr<int>(); works with VC6.0 but elicits the [AFAICT] uninformative <emsg>parse error before `;'</emsg> from gcc version 2.96 20000731 (Red Hat Linux 7.3 2.96-110). Drat! "Duane Murphy" <duanemurphy@mac To: "Boost Users" <Boost-Users(a)yahoogroups.com> .com> cc: Subject: RE: [Boost-Users] NRVO (was Re: "Correct" way to return empty shared_ptr?) 04/24/2003 09:40 AM Please respond to Boost-Users --- At Thu, 24 Apr 2003 16:56:40 +0100, Ben Hutchings wrote: <snip> shared_ptr<int> maybe_get_a_pointer() { return shared_ptr<int>(); } is perfectly reasonable and can also be optimized. Is there more recent and complete discussion concerning this optimization? Is naming actually required? ...Duane Info: <http://www.boost.org> Wiki: <http://www.crystalclearsoftware.com/cgi-bin/boost_wiki/wiki.pl> Unsubscribe: <mailto:boost-users-unsubscribe@yahoogroups.com> Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/

1 0

RE: [Boost-Users] Re: pool_allocator problems on VC7.1 (real rele ase)
by scleary＠jerviswebb.com 25 Apr '03

25 Apr '03

> -----Original Message----- > From: John Harris [mailto:john.harris@tradingtechnologies.com] > > Since it seems to work on VC7.1, it would be nice if the author had > access to that compiler and could do some testing. In the mean time, > I'll try some stuff with this new toy and report any further problems. Some command-line tools ship with their DDK, so I do have a version of "cl.exe" that reports "13.00.9176" - no idea how this fits in with the VC versions, though. And the DDK doesn't include an STL anyway. There is one "common" failure case with some implementations of STL containers: passing 0 to Allocator::deallocate. When used with std::allocator, the 0 is passed through to operator delete() which of course ignores it. When used with pool_alloc, this generally causes a segfault. IIRC, STLPort's std::vector will do this if you construct & destroy an empty vector. Since I don't have the VC STL, could you do that test case for me? Thanks, -Steve

2 2

MP3 and C++
by steelgig 25 Apr '03

25 Apr '03

Hi, I'm working on a project for the university and I'm looking for some C++ library or application that could help me. My goal is to extract some kind of information related to the spectra of a signal coded in MP3. I'm looking for a spectrum analyzer that localizes peaks in the power spectrum (better if it works on a user-specified band) and allows the detection of the temporal occurence of each peak and its intensity. I also need to retrieve some sample of frequency components (or the MDCT coefficients) near the identified peaks. Every suggestion is welcome (even something that satisfies only partially the requests or that indicates a better place to found an answer to this message). Thanks in advance for the help, Gio PS: even some function that works on the MP3 file without decoding it, for example only partial decoding actions (IMDCT, Huffman decoding) or subband value retrieval function, could be useful.

1 0

Circular Buffer
by Joe Mariadassou 24 Apr '03

24 Apr '03

Does boost or any other open source have an implementation of a bounded circular buffer? My search in boost and google yielded naught. The main purpose of this container is to pipeline messages in a typical producer-consumer scenario. I was surprised not to find any implementation. TIA Joe [Non-text portions of this message have been removed]

2 1

Test for endian using MPL or PP(Wave?)
by Mike Wilson 24 Apr '03

24 Apr '03

Hi, There is an application I use that always produces a little endian binary file. I have written a program to process this file. As my program can port to both little/big endian platforms I have a #define (LITTLE_ENDIAN_TARGET) acting as a flag for the necessary byte swapping that is needed on the big-endian platforms. I was wondering if there is a better way to do this, so I searched the Web & boost archives. I found the following code snippet in the boost archives. bool isLittleEndian() { int temp = 1; return (*(char*)&temp) != 0; } which is of some help (eliminating my #define). However, I was wondering if there is a PP or mpl way of testing for endian-ness at compile time. For example, I'd like to do something like the following: const bool IsLittleEndian = ..to be determined automatically..; I've got a gut feeling that I can't do the above at complile time, but maybe something like Wave may do the job? Thanks for any assistance in this matter. Mike

4 4

RE: [Boost-Users] Re: "Correct" way to return empty shared_ptr?
by Ben Hutchings 24 Apr '03

24 Apr '03

dick.bridges(a)tais.com wrote: > This is one of the places where I get *REALLY* confused - the 'real' code > has multiple returns. Apologies: I should have included this psuedo code > in the original example. It looks like this. > > <snippet> > // v is populated somewhere else > typedef vector<shared_ptr<int> > V; > V v; > > shared_ptr<int> maybe_get_a_pointer() > { > shared_ptr<int>::iterator itr; > itr = find_if(v.begin(), v.end(), some_op); > > if(itr != v.end) > return(*itr); > > shared_ptr<int> p; > return(p); > } > </snippet> > > Are compilers supposed to be able to match *itr to p in order to apply the > NRVO? I don't know what you mean by "match *itr to p". The NRVO should be possible for the second return statement, because if p is constructed then it's always returned. However, optimisations are by definition optional so the implementation isn't "supposed to" perform them.

1 0

RE: [Boost-Users] NRVO (was Re: "Correct" way to return emptyshared_ptr?)
by Ben Hutchings 24 Apr '03

24 Apr '03

Duane Murphy <duanemurphy(a)mac.com> wrote: > --- At Thu, 24 Apr 2003 16:56:40 +0100, Ben Hutchings wrote: > >dick.bridges(a)tais.com wrote: > >> I don't understand C++ well enough to grok a compiler's interpretation of > >> the code, but I seems like I'm creating an extra copy for the return. > ><snip> > > > >Logically the local variable is copied to the the caller, but since the type > >of the returned expression matches the return type the compiler is allowed > >to optimise away that copy, even though the return type has a user-defined > >copy constructor. In this case you always return a certain named variable, > >so the compiler can arrange to construct it in the memory location where the > >caller expects the returned value. This is called the named-return-value > >optimisation, or NRVO. > > I dont want to start any kind of war, and this is a bit off topic, but I > was investigating NVRO and found a few discussions with Scott Meyers that > indicated that naming is not relevant to the optimization. That this: > > shared_ptr<int> maybe_get_a_pointer() > { > return shared_ptr<int>(); > } > > is perfectly reasonable and can also be optimized. > > Is there more recent and complete discussion concerning this > optimization? Is naming actually required? I realise that the implementation is allowed to optimise out the copy whether the return value is named or not. Actually doing so requires quite different treatment of the two cases, so it's worth considering them as separate optimisations. For NRVO to be possible at some return statement, there must not be any return statements reachable after the construction of the returned object that return any other object. The optimisation involves changing the code generated for the definition of the object, rather than the code for the return statement. So the implementation has to consider the whole function when doing this. Anonymous RVO is a very local optimisation, so it should require less work to apply.

1 0

RE: [Boost-Users] Re: "Correct" way to return empty shared_ptr?
by dick.bridges＠tais.com 24 Apr '03

24 Apr '03

This is one of the places where I get *REALLY* confused - the 'real' code has multiple returns. Apologies: I should have included this psuedo code in the original example. It looks like this. <snippet> // v is populated somewhere else typedef vector<shared_ptr<int> > V; V v; shared_ptr<int> maybe_get_a_pointer() { shared_ptr<int>::iterator itr; itr = find_if(v.begin(), v.end(), some_op); if(itr != v.end) return(*itr); shared_ptr<int> p; return(p); } </snippet> Are compilers supposed to be able to match *itr to p in order to apply the NRVO? "Ben Hutchings" <ben.hutchings@businesswebso To: <Boost-Users(a)yahoogroups.com> ftware.com> cc: Subject: RE: [Boost-Users] Re: "Correct" way to return empty shared_ptr? 04/24/2003 08:56 AM Please respond to Boost-Users dick.bridges(a)tais.com wrote: > I don't understand C++ well enough to grok a compiler's interpretation of > the code, but I seems like I'm creating an extra copy for the return. <snip> Logically the local variable is copied to the the caller, but since the type of the returned expression matches the return type the compiler is allowed to optimise away that copy, even though the return type has a user-defined copy constructor. In this case you always return a certain named variable, so the compiler can arrange to construct it in the memory location where the caller expects the returned value. This is called the named-return-value optimisation, or NRVO. Info: <http://www.boost.org> Wiki: <http://www.crystalclearsoftware.com/cgi-bin/boost_wiki/wiki.pl> Unsubscribe: <mailto:boost-users-unsubscribe@yahoogroups.com> Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/

1 0

RE: [Boost-Users] Re: "Correct" way to return empty shared_ptr?
by Ben Hutchings 24 Apr '03

24 Apr '03

dick.bridges(a)tais.com wrote: > I don't understand C++ well enough to grok a compiler's interpretation of > the code, but I seems like I'm creating an extra copy for the return. <snip> Logically the local variable is copied to the the caller, but since the type of the returned expression matches the return type the compiler is allowed to optimise away that copy, even though the return type has a user-defined copy constructor. In this case you always return a certain named variable, so the compiler can arrange to construct it in the memory location where the caller expects the returned value. This is called the named-return-value optimisation, or NRVO.

2 1

Re: [Boost-Users] Re: "Correct" way to return empty shared_ptr?
by dick.bridges＠tais.com 24 Apr '03

24 Apr '03

I don't understand C++ well enough to grok a compiler's interpretation of the code, but I seems like I'm creating an extra copy for the return. Oh well, if the snippet seems ok, far be it from me to throw asparagus on it. ;) Thanks for the help. "Edward Diener" <eddielee@tropic To: boost-users(a)yahoogroups.com soft.com> cc: Sent by: news Subject: [Boost-Users] Re: "Correct" way to return empty shared_ptr? <news(a)main.gmane .org> 04/23/2003 05:47 PM Please respond to Boost-Users dick.bridges(a)tais.com wrote: > What is the 'correct' way to return a NULL/empty shared_ptr? For > example, this works > > <snippet> > shared_ptr<int> maybe_get_a_pointer() > { > shared_ptr<int> p; > return(p); > } > </snippet> > > but seems 'klunky'. I've clearly missed something in the > docs/examples somewhere. Can someone point me in the right direction? What is klunky about it ? It seems pretty normal to me. Another other possibility is to passed a reference to a shared_ptr<> and have your function above fill it in with an actual pointer to int and return a bool true if it succeeded or false if it did not. Another possibility is to use boost::optional<>. Another one to pass back a bool,shared_ptr<> std::pair or boost::tuple. But I prefer the method you used originally. Info: <http://www.boost.org> Wiki: <http://www.crystalclearsoftware.com/cgi-bin/boost_wiki/wiki.pl> Unsubscribe: <mailto:boost-users-unsubscribe@yahoogroups.com> Your use of Yahoo! Groups is subject to http://docs.yahoo.com/info/terms/

1 0

uBlas Lapack binds from Sandbox - VC6 workaround
by geert_ceuppens 24 Apr '03

24 Apr '03

As an introduction; my goal is to link together the uBlas matrix/vector classes with the Intel MKL library (I looked at ATLAS, but the Lapack coverage was insufficient). I have tried the bindings code in the Boost Sandbox (on which Toon and Kresimir are working), but VC6 falls over. I managed to work around the partial template specialisation problem (see my previous postings), thanks to Kresimir's and Toon's feedback. There is however one remaining question I need some help with. First the question; after that I post all the code which I've hacked together to assist other interested users using VC6. Due to my ignorance, I am obliged to pass the orientation as a template parameter to the matrix_traits class. What I would like to do is to obtain this information from the matrix class directly. The issue is that matrix_type::functor_type does not exist for all matrix classes. For example, hermitian, triangular and symmetrix matrix use matrix_type::functor2_type to store the orientation of the matrix. With a good compiler (i.e. *NOT* VC6), one could specialise matrix_traits for these matrix classes (at least that is what I think - but this isn't done yet in the ublas_matrix.hpp file in the Sandbox), here however I'm stuck. As a hack, I allow the orientation to be passed as a template parameter. Is there a better way to do this (the matrix class M knows the orientation already)? // M is type of matrix, RC is the row/column orientation template <typename M, // type of matrix typename RC = boost::numeric::ublas::row_major> struct matrix_traits { public: typedef RC row_column_major; typedef M matrix_type; ... static int stride1 (matrix_type& m) { //return matrix_type::functor_type::one1 (m.size1(), m.size2 ()); return row_column_major::one1 (m.size1(), m.size2()); } ... }; Here is the code I wrote: ---------------------//---------------------- traits.h: // (-*- C++ -*- header) /* * * Copyright (c) Kresimir Fresl and Toon Knapen 2002 * * Permission to copy, modify, use and distribute this software * for any non-commercial or commercial purpose is granted provided * that this license appear on all copies of the software source code. * * Authors assume no responsibility whatsoever for its use and makes * no guarantees about its quality, correctness or reliability. * * First author acknowledges the support of the Faculty of Civil * Engineering, University of Zagreb, Croatia. * * --> Modified/simplified for VC6 */ #ifndef BOOST_NUMERIC_BINDINGS_TRAITS_TRAITS_HPP #define BOOST_NUMERIC_BINDINGS_TRAITS_TRAITS_HPP #include <cstddef> //namespace boost { namespace numeric { namespace bindings { namespace traits { // vector_traits<> base: // .. implementation -- do not use it directly template <typename V, typename T = typename V::value_type> struct default_vector_traits { typedef T value_type; typedef T* pointer; // assumption: iterator==pointer // .. e.g. ublas::(un)bounded_array static pointer storage (V& v) { return v.begin(); } static const T* storage (const V& v) { return v.begin(); } static int size (V& v) { return v.size(); } static int size (const V& v) { return v.size(); } static int stride (V& v) { return 1; } static int stride (const V& v) { return 1; } }; // vector_traits<> generic version: template <typename V> struct vector_traits : default_vector_traits<V> {}; // matrix structure tags: struct general_t {}; struct symmetric_t {}; struct symmetric_packed_t {}; struct hermitian_t {}; struct hermitian_packed_t {}; // triangular, banded etc. struct unknown_structure_t {}; /* // matrix_traits<> generic version: template <typename M> struct matrix_traits { // typedefs: // matrix_structure // value_type // pointer // ordering_type // static functions: // pointer storage() // int storage_size() // int size1() // int size2() // int leading_dimension() // not all matrix types // symmetric/hermitian typedefs: // uplo_type }; */ // storage ordering tags: struct row_major_t {}; struct column_major_t {}; // upper/lower tags: struct upper_t {}; struct lower_t {}; // free accessor functions: template <typename VectorT> inline typename vector_traits<VectorT>::pointer vector_storage (VectorT& v) { return vector_traits<VectorT>::storage (v); } template <typename VectorT> inline int vector_size (VectorT& v) { return vector_traits<VectorT>::size (v); } template <typename VectorT> inline int vector_stride (VectorT& v) { return vector_traits<VectorT>::stride (v); } // All the matrix_traits<MatrixT> code has been removed here ... }}}} #endif // BOOST_NUMERIC_BINDINGS_TRAITS_TRAITS_HPP ---------------------//---------------------- ublas_matrix.h: // (-*- C++ -*- header) /* * * Copyright (c) Kresimir Fresl and Toon Knapen 2002 * * Permission to copy, modify, use and distribute this software * for any non-commercial or commercial purpose is granted provided * that this license appear on all copies of the software source code. * * Authors assume no responsibility whatsoever for its use and makes * no guarantees about its quality, correctness or reliability. * * First author acknowledges the support of the Faculty of Civil * Engineering, University of Zagreb, Croatia. * * --> Modified/simplified for VC6 */ #ifndef BOOST_NUMERIC_BINDINGS_TRAITS_UBLAS_MATRIX_H #define BOOST_NUMERIC_BINDINGS_TRAITS_UBLAS_MATRIX_H #include <boost/numeric/ublas/matrix.hpp> #include "traits.h" namespace boost { namespace numeric { namespace bindings { namespace traits { // Need to find a way to extract the orientation class (row / column) template <typename M, // type of matrix typename RC = boost::numeric::ublas::row_major> // row or column major struct matrix_traits { public: typedef RC row_column_major; typedef general_t matrix_structure; typedef M::value_type value_type; typedef M::value_type *pointer; typedef M matrix_type; static pointer storage (matrix_type& m) { typedef typename matrix_type::array_type array_type; return vector_traits<array_type>::storage (m.data()); } static const value_type *storage (const matrix_type& m) { typedef typename matrix_type::array_type array_type; return vector_traits<array_type>::storage(m.data()); } static int size1 (matrix_type& m) { return m.size1(); } static int size1 (const matrix_type& m) { return m.size1(); } static int size2 (matrix_type& m) { return m.size2(); } static int size2 (const matrix_type& m) { return m.size2(); } static int storage_size (matrix_type& m) { return size1 (m) * size2 (m); } static int storage_size (const matrix_type& m) { return size1 (m) * size2 (m); } // stride1 == distance (m (i, j), m (i+1, j)) static int stride1 (matrix_type& m) { //return matrix_type::functor_type::one1 (m.size1(), m.size2 ()); return row_column_major::one1 (m.size1(), m.size2()); } static int stride1 (const matrix_type& m) { //return matrix_type::functor_type::one1 (m.size1(), m.size2 ()); return row_column_major::one1 (m.size1(), m.size2()); } // stride2 == distance (m (i, j), m (i, j+1)) static int stride2 (matrix_type& m) { //return matrix_type::functor_type::one2 (m.size1(), m.size2 ()); return row_column_major::one2 (m.size1(), m.size2()); } static int stride2 (const matrix_type& m) { //return matrix_type::functor_type::one2 (m.size1(), m.size2 ()); return row_column_major::one2 (m.size1(), m.size2()); } static int leading_dimension (matrix_type& m) { //return matrix_type::functor_type::size2 (m.size1(), m.size2 ()); return row_column_major::size2 (m.size1(), m.size2()); } static int leading_dimension (const matrix_type& m) { //return matrix_type::functor_type::size2 (m.size1(), m.size2 ()); return row_column_major::size2 (m.size1(), m.size2()); } }; }}}} #endif ---------------------//---------------------- boost_mkl_lapack.cpp: namespace ublas = boost::numeric::ublas; #ifdef BOOST_MSVC #pragma warning (disable: 4355) #pragma warning (disable: 4503) #pragma warning (disable: 4786) #endif #define BOOST_UBLAS_USE_EXCEPTIONS #include <iostream> #include <boost/numeric/ublas/matrix.hpp> #include <boost/numeric/ublas/symmetric.hpp> #include <vector> // Intel MKL Lapack #include <mkl/include/mkl_lapack64.h> #include "ublas_matrix.h" // To make life easiertypedef ublas::matrix<double,ublas::row_major> Matrix; typedef std::vector<int> IntVector; template<class M> int getrf(M& a, IntVector& ipiv) { typedef matrix_traits<M> mtraits; typedef vector_traits<IntVector> vtraits; int m = a.size1(); int n = a.size2(); assert(ipiv.size() == (m < n ? m : n)); double *aStorage = mtraits::storage(a); int lda = mtraits::leading_dimension(a); int *ipivStorage = vtraits::storage(ipiv); int info; // return status dgetrf(&m, &n, aStorage, &lda, ipivStorage, &info); return info; } template<class M> int getrs(const M& a, const IntVector& ipiv, M& b, char trans = 'N' ) { typedef matrix_traits<M> mtraits; typedef vector_traits<IntVector> vtraits; int m = a.size1(); int n = a.size1(); assert(ipiv.size() == (m < n ? m : n)); // Note: we need to cast away the const for a and ipiv double *aStorage = const_cast<double*>(mtraits::storage(a)); double *bStorage = mtraits::storage(b); int lda = mtraits::leading_dimension(a); int ldb = mtraits::leading_dimension(b); int *ipivStorage = const_cast<int*>(vtraits::storage(ipiv)); // if b is column major, take size2 (i.e. columns) int nrhs = mtraits::size1(b); int info; // return status dgetrs(&trans, &n, &nrhs, aStorage, &lda, ipivStorage, bStorage, &ldb, &info); return info; } template <typename M> void init_symm (M& m, char uplo = 'f') { size_t n = matrix_traits<M>::size1 (m); for (size_t i = 0; i < n; ++i) { m(i, i) = n; for (size_t j = i + 1; j < n; ++j) { if (uplo == 'u' || uplo == 'U') m(i, j) = n - (j - i); else if (uplo == 'l' || uplo == 'L') m(j, i) = n - (j - i); else m(i, j) = m(j, i) = n - (j - i); } } } // printing: template <typename M> void print_m (const M& m, char const* ch = 0) { if (ch) std::cout << ch << ":\n"; size_t sz1 = matrix_traits<M>::size1 (m); size_t sz2 = matrix_traits<M>::size2 (m); for (std::size_t i = 0 ; i < sz1 ; ++i) { for (std::size_t j = 0 ; j < sz2 ; ++j) std::cout << m(i, j) << " "; std::cout << std::endl; } // std::cout << std::endl; } // printing: template <typename V> void print_v (V const& v, char const* ch = 0) { if (ch) std::cout << ch << ": "; size_t sz = vector_traits<V>::size (v); for (std::size_t i = 0; i < sz; ++i) std::cout << v[i] << " "; std::cout << std::endl; } using std::size_t; void main(void) { size_t n = 3; Matrix a(n, n), acopy(n, n); IntVector ipiv(n); init_symm(a); acopy = a; print_m (a, "A"); // Now factor int status = getrf(a, ipiv); print_v(ipiv, "ipiv"); print_m (a, "A"); // The solution (and the right hand side) are row oriented size_t nrhs = 2; Matrix b(nrhs, n), x(nrhs, n); int i; for(i = 0; i < n; i++) { x(0,i) = (double)i + 1; x(1,i) = (double)i + 4.0; } b = ublas::trans(ublas::prod(acopy, ublas::trans(x))); print_m (b, "b original"); print_m (x, "x original"); status = getrs (a, ipiv, b); // solve from factorization print_m (b, "X"); std::cout << std::endl; // Now try the same for a symmetrix matrix ublas::symmetric_matrix<double, ublas::lower> ml(n, n); init_symm(ml); print_m(ml, "symmetric matrix"); status = getrf(ml, ipiv); print_v(ipiv, "ipiv"); print_m (a, "A"); } You will need your project settings to be such that you link with the Intel lib files and the DLL should be in your path (see Intel MKL documentation). Your link tab in project/settings should contain (the lib directory is set using tools/options, same for the include): mkl_c.lib mkl_p3.lib mkl_lapack.lib You may want to do this in a different namespace than the boost namespace (I actually commented out all the namespace stuff above, but reintroduced it in this posting - if it does not compile, than that's probably the reason). More binds to Lapack still need to be done, but this is a start already. Geert

2 1

Intel compiler (7.0), STLport (4.3.5) and Boost?
by Andre du Toit 24 Apr '03

24 Apr '03

Is it possible to build bjam and the boost libraries using the Intel compiler (7.0) in conjunction with STLport(4.3.5) under Windows (2000)? If not, please tell me and do not bother reading the rest of this email! I did manage to build bjam and the boost libraries using msvc-stlport toolset, but we use the Intel compiler in conjunction with STLport for our Windows applications. Now I know I can't just start using the Intel compiler with source using boost libraries which were built with a different toolset (or could I?). I get a number of errors regarding multiple definitions of ptime iostream operators. But before I present the exact error messages, I would like to clear up the Intel + STLport issue. I would appreciate any suggestions. Greetings, Andre.

1 0

VC7 internal compiler error in mpl using graph & spirit
by Julius Muschaweck 24 Apr '03

24 Apr '03

Hi, with VC 7.0, with both boost_1_30 and the current CVS, and with both release and debug mode, #include <boost/graph/adjacency_list.hpp> #include <boost/spirit/core.hpp> gives an Internal Compiler Error 1001 at line 52 of h:\jmj\C++\Include\boost_1_30_0\boost\mpl\aux_\preprocessed\msvc70\apply.hpp while the error disappears when changing the order of the include files. I can live with the workaround of course, but maybe someone in charge of mpl would like to go after this. Regards, Julius

1 0

RE: [Boost-Users] Sorting list < shared_ptr < foo > > using std::list::sort()
by Ben Hutchings 24 Apr '03

24 Apr '03

heath_r_davis <1heathdavis(a)msn.com> wrote: > I have experienced some difficulty with sorting lists of type: > > list < shared_ptr < foo > > <snip> You need to specify your own ordering predicate, because shared_ptr<T>::operator< does *not* call T::operator< . Also remember that if operator< is a member function it should be const.

1 0

uBlas matrix - storage class
by geert_ceuppens 24 Apr '03

24 Apr '03

Is there a performance difference between the different storage classes for matrix and vector? If so, which one is the most performant? I could not find any reference to this in the documentation ... :^( Many thanks - Geert

2 1

"Correct" way to return empty shared_ptr?
by dick.bridges＠tais.com 24 Apr '03

24 Apr '03

What is the 'correct' way to return a NULL/empty shared_ptr? For example, this works <snippet> shared_ptr<int> maybe_get_a_pointer() { shared_ptr<int> p; return(p); } </snippet> but seems 'klunky'. I've clearly missed something in the docs/examples somewhere. Can someone point me in the right direction? TIA

2 1

RE: [Boost-Users] Re: pool_allocator problems on VC7.1 (real release)
by Ben Hutchings 23 Apr '03

23 Apr '03

John Harris <john.harris(a)tradingtechnologies.com> wrote: > Ben Hutchings wrote: > > I compiled almost the exact same code: > > > > #include <boost/pool/pool_alloc.hpp> > > #include <map> > > #include <utility> > > > > typedef std::basic_string<char, std::char_traits<char>, > > boost::pool_allocator<char> > t; > > > > int main() > > { > > std::map<t, t, std::less<t>, > > boost::pool_allocator<std::pair<const t, t> > > m; > > } > > > > with VC++ 7.0 and Boost 1.29, and got no warnings or errors at all. > > > > Ah, yes, that works for me, too (no warning, either), but try adding > a line of code that actually uses the map, like this: > > m["abc"] = "def"; > > and you get the following cascade of errors: <snip> Which all indicates that operator< is not defined for t, because we didn't include <string>. Let's not be so quick to blame the compiler. VC++ 7 is a long way ahead of version 6.

2 1

Sorting list < shared_ptr < foo > > using std::list::sort()
by heath_r_davis 23 Apr '03

23 Apr '03

I have experienced some difficulty with sorting lists of type: list < shared_ptr < foo > > Here is the expected (sorted) program output: 424238335 596516649 719885386 846930886 1189641421 1649760492 1681692777 1714636915 1804289383 1957747793 Instead, I end up with: 1804289383 846930886 1681692777 1714636915 1957747793 424238335 719885386 1649760492 596516649 1189641421 Below is the code in question: //////////////////////////////////////////////// #include <iostream> #include <list> using namespace std; #include <boost/shared_ptr.hpp> using namespace boost; //////////////////////////////////////////////// class my_class { public: int m_val; public: my_class(); bool operator<(const my_class&); void Display(); }; //////////////////////////////////////////////// my_class::my_class() { m_val = rand(); } bool my_class::operator<(const my_class& arg) { return m_val < arg.m_val; } void my_class::Display() { cout << m_val << "\n"; } //////////////////////////////////////////////// int main( int argc, char **argv ) { // list < my_class > my_list; // list < my_class >::iterator my_iter; list < shared_ptr < my_class > > my_list; list < shared_ptr < my_class > >::iterator my_iter; //populate my_list for (int x=0; x<10; x++) { // my_class tmp_obj; shared_ptr < my_class > tmp_obj; tmp_obj.reset(new my_class); my_list.push_back(tmp_obj); } //sort the list my_list.sort(); //display contents of my_list my_iter = my_list.begin(); while (my_iter != my_list.end()) { // (*my_iter).Display(); (*my_iter)->Display(); ++my_iter; } return 0; } //////////////////////////////////////////////// Thanks in advance for any help you may render...

1 0

RE: [Boost-Users] Problems linking with Boost 1.29.0 thread lib, Linux/g++ 3.2.2 -r2
by Ben Hutchings 23 Apr '03

23 Apr '03

subsample <knuarv(a)nith.no> wrote: > Failing to link a program with libboost_thread, even though I supply > the -lboost_thread directive in the makefile. I'm kind of new to > makefiles, having only used Visual C++ in Windows untill recently. I > 've tried changing things around a bit, doesn't help. A minimal test > program I wrote links fine with boost_thread though. My makefile looks > like this: > CC=g++ > CFLAGS=-g > LDFLAGS=-L/home/sub/Libs/alsa/lib -lasound -lboost_thread -lpthread > > alsatest: alsatest.o > $(CC) $(LDFLAGS) -o alsatest SmallObj.o alsatest.o <snip> ld doesn't work this way. -l is just a shortcut for finding a library file. The libraries specified with -l are part of the file list, same as the object files and any library files specified by their full file name. The nasty thing is that backwards references are only resolved in object files, not in library files, so library files must be listed after the object files (and any other library files) that reference them. You need to take the libraries from LDFLAGS (perhaps into another variable) and put them at the end of the command line. Also you should be using CXX and CXXFLAGS for compiling C++ code, not CC and CFLAGS. Since you're building a pthreads program, you *must* define the macro _REENTRANT to get a threadsafe C library. As a shortcut, include -pthread in CXXFLAGS (this causes linking with -lpthread, too).

2 1

RE: [Boost-Users] pool_allocator problems on VC7.1 (real release)
by Ben Hutchings 23 Apr '03

23 Apr '03

John Harris <john.harris(a)tradingtechnologies.com> wrote: > scleary(a)jerviswebb.com wrote: > > John - > > > > Sorry, Boost.Pool does not currently support MSVC, due to that compiler's > > extreme lack of template support. > > > > I also tried this on the real released VC7.1 (which has been on MSDN > subscriber downloads for a couple of weeks, now), and, while it > compiles, I get warnings that point me to some code that looks like > it's losing precision (size_type reduced to bool) in a calculation. VC++ has produced that warning for a while now, and it's totally bogus. Notice that it's a "performance warning" not an "are you sure this is that you want?" warning, and it's being generated here for *casts* not implicit conversions. <snip> > Another test I tried was a map<string,string>, actually declared like > this: > > typedef > std::basic_string<char,std::char_traits<char>,boost::pool_allocator<ch > ar> > t; > std::map<t, t, std::less<t>, boost::pool_allocator<std::pair<const t, > t> > > m; > > ...and got these errors: <snip> I compiled almost the exact same code: #include <boost/pool/pool_alloc.hpp> #include <map> #include <utility> typedef std::basic_string<char,std::char_traits<char>,boost::pool_allocator<char > > t; int main() { std::map<t, t, std::less<t>, boost::pool_allocator<std::pair<const t, t> > > m; } with VC++ 7.0 and Boost 1.29, and got no warnings or errors at all. The command line was: cl /D__WIN32__ /EHsc /I"c:\program files\boost" test.cpp I repeated this with Boost 1.30 and again got no warnings or errors. I don't have VC++ 7.1 to hand, but I would be surprised if it broke anything that worked in 7.0.

2 1

Signal/Slot copy semantics.
by Jesper Bojesen 23 Apr '03

23 Apr '03

I have a problem with the sematics of the the signal connect method. Consider this simple program: #include <iostream> #include <boost/signal.hpp> struct HelloWorld { void operator()() const { std::cout << "Hello, World!, the parallel world exist at " << (void *) this << std::endl; } }; int main () { // Connect a HelloWorld slot HelloWorld hello; std::cout << "Main world at " << (void *) &hello << std::endl; boost::signal<void ()> sig; sig.connect(hello); sig (); std::cout << "Thats all folks!\n"; return 0; } It produces this output: >Main world at 0xbffff2e0 >Hello, World!, the parallel world exist at 0x8056128 >Thats all folks! My problem is that the object signalled is not the same object that I subscribed to the signal. Is this really what was inteded ? Is the any rationale behind this behaviour ? -- Jesper Bojesen Medical Insight Email: jbb(a)medical-insight.com Phone mobile: (+45) 40 99 55 03 Phone direct: (+45) 46 55 04 42 Fax: (+45) 46 55 11 66

3 5

UBLAS and algorithms
by David Johnson 23 Apr '03

23 Apr '03

Hi, I'm looking to solve some matrix problems using conjugate gradient methods. CGS and GMRES to name a couple. I see where the matricies can be stored by the UBLAS classes, but I don't see any of the algorithms available in boost. Are they available somewhere? I've looked at packages like IML++ and see them there, but they appear to use different storage classes. Can anyone point me in the direction of where some boost compatible algorithms are? Thanks, Dave

1 0

RE: pool_allocator problems on VC7.1 (real release)
by scleary＠jerviswebb.com 23 Apr '03

23 Apr '03

> -----Original Message----- > From: John Harris [mailto:john.harris@tradingtechnologies.com] > I also tried this on the real released VC7.1 (which has been on MSDN > subscriber downloads for a couple of weeks, now), and, while it > compiles, I get warnings that point me to some code that looks like > it's losing precision (size_type reduced to bool) in a calculation. > The pool code about which the following warnings complain: > > const size_type num_chunks = total_req_size / partition_size + > static_cast<bool>(total_req_size % partition_size); > The cast is intentional. The intent is to add one if the result of the modulus is non-zero. I don't know why the compiler is producing warnings, since I'm using an explicit cast. > Another test I tried was a map<string,string>, actually declared like > this: I'm having a hard time deciphering the compiler errors (I'm not used to MSVC errors), but it looks like it might be a problem in the STL. Have you tried this with STLPort? > If you have an example of a map<string,string> using pool_allocator > that compiles on VC7.1, I'd *love* to see it. That's all I'm really > after, here. Is MSVC available for download for non-MSDN subscribers? If so, I can download it and try to get it working. -Steve

2 1

Trouble grokking iterator adaptors
by Kris Braun 23 Apr '03

23 Apr '03

Hi there, I could really use the features provided by iterator adaptors part of the library, but I'm having a lot of trouble getting a nested set of adaptors to compile. In one case, I'd like to return an iterator into std::map< std::string, boost::shared_ptr< ApiOption > >, and have it return references to options in which option->isTemplate( ) is false. Conceptually, I'd like a filter_iterator->indirect_iterator- >projection_iterator chain. I've been using this typedef to project onto the second member of the pair: typedef boost::projection_iterator_pair_generator< select_second< std::string, boost::shared_ptr< ApiOption > >, OptionMap::iterator, OptionMap::const_iterator > OptionMapIterators; but I've been at a loss to add indirect_iterator and filter_iterator into the mix. Can anyone help me get unstuck? Thank you in advance, Kris

5 7