lapackpp-devel — Discussion of Lapack++ development

[Lapackpp-devel] Getting Link error when building 64 bit version

[<su...@it...>]

Hello,

   I am getting following link error when building 64 bit version.
 
   Error    39    error LNK2019: unresolved external symbol dasum_ 
referenced in function "double __cdecl Blas_Norm1(class LaVectorDouble 
const &)" (?Blas_Norm1@@YANAEBVLaVectorDouble@@@Z)    blas1pp.obj

   I think reason for it , there is no 64 bit version of  
liblapack32.lib and libblas32.lib.
   I was not having source code to build this two in 64 bit version.

   Can anybody help me to get these libraries or any other way to remove 
the linking error.

Thanks
Sumant

[Lapackpp-devel] How to compile a release version of lapack dll in MS C++ 2008

[Yassine K. <yas...@gm...>]

Hi,

I have successfully compiled and tested lapack++ on MSVC++ 2008.
However, I noticed that the lapack MSVC project (downloaded from
lapack web site) can only be compiled in Debug mode. Is there a
project that can be complied in Release mode (to obtain the
lapack.dll) so I can distribute it along with my application? Any
other help would be appreciated.

Best regards,
-Yassine

Re: [Lapackpp-devel] Lapack++ and parallel computing

[Christian S. <sti...@tu...>]

Am Mittwoch, 27. Januar 2010 schrieb Louis Ixo:
> Dear all,
> 
> I'm new to lapack++ and I would like to know if lapack++ is a parallel
> computing library that can uses something like OpenMP or etc... for use
> multiple-CPUs calculations, or maybe is there a way to activate it? Did I
> need to build blas and lapack with openmp before build lapack++?

Unfortunately no: Lapack++ has no means of using the potentials of parallel 
computing. Even if blas and lapack had been built with openmp (if that were 
possible), lapack++ won't be able to use any advantages from that because 
nobody ever tried to improve it in that direction. Sorry for that.

Regards,

Christian Stimming

[Lapackpp-devel] Lapack++ and parallel computing

[Louis I. <li...@ho...>]

Dear all,

I'm new to lapack++ and I would like to know if lapack++ is a parallel
computing library that can uses something like OpenMP or etc... for use
multiple-CPUs calculations, or maybe is there a way to activate it? Did I
need to build blas and lapack with openmp before build lapack++?
Thank you very much for any kind of explanation.
Best Regards,
L. Ixo

Re: [Lapackpp-devel] Lapack++ on Win64

[Christian S. <sti...@tu...>]

Dear Farshid,

this question should have been sent to the lapackpp-devel mailing list (now in 
CC), but never mind.

This particular issue has been a request by several people by now. I thought 
someone would already commit this into SVN, but apparently this hasn't 
happened so far.

If you have something which you think you can contribute, just send it to the 
mailing list and I would happily include it in our SVN and the next releases.

Regards,

Christian

Am Freitag, 18. Dezember 2009 schrieb Farshid Mossaiby:
> Dear Christian,
> 
> I have been trying for a while to find a way to compile Lapack++ on Win64
>  platform with MSVC 2008. I could finally do it, and used AMD ACML BLAS /
>  LAPACK to do this. The test (tgd) ran fine.
> 
> While I was trying to compile the project, I had to create new
>  configurations from old ones, which made some problems, i.e. MSVS thought
>  the project is a Makefile type project. After correcting it, and setting
>  appropriate switches, everything ran smooth.
> 
> May I suggest adding Win64 platform to your MSVC solution? This will make
>  compiling Lapack++ for Win64 much easier. I can help if you need.
> 
> Best regards,
> Farshid Mossaiby
> 

Re: [Lapackpp-devel] Question regarding Lapack++ in MSVC 6.0

[Christian S. <sti...@tu...>]

Zitat von Yassine Kabouzi <yas...@gm...>:

> Hello Christian,
>
> I am interested in using Lapack++ for a project that requires singular
> value decomposition. The problem I am having is that I need to compile
> a dll in MSVC 6.0.
>
> Is there any old version that I can compile with MSVC 6.0 and will do
> well for SVD?

I don't think any of the lapackpp versions will compile with MSVC 6.0  
as there is some non-trivial use of template classes, which the old  
MSVC probably doesn't support.

Regards,

Christian

Re: [Lapackpp-devel] Matrixaddition in Lapack++

[Christian S. <sti...@tu...>]

Zitat von Henning Lategahn <Hen...@ki...>:
> Zur Zeit verewende ich die Bibliothek Lapack++, die sie geschrieben  
> haben. Soweit bin ich sehr zufrieden damit. Jedoch weiss ich nicht  
> wie ich zwei LaGenMatDouble Matricen addieren kann. Ich bin also auf  
> der Suche nach dem Matrixäquivalent zu Blas_Add_Mult. Wenn ich  
> Matrizen elementweise addiere (über verschachtelte for-schleifen)  
> sagt mir mein Profiler, dass ich sehr viel Zeit in operator()(int,  
> int) verbringe.

Zwei Möglichkeiten: 1. Der operator+, also

Matrix A = B + C;
http://lapackpp.sourceforge.net/html/blaspp_8h.html

2. "Missbrauchen" von Blas_Mat_Mat_Mult mit einer dritten Einheitsmatrix:
http://lapackpp.sourceforge.net/html/blas3pp_8h.html#044bbe029632ffac36a894b1071cb22f

I=LaGenMatDouble::eye(A.size(1));
Blas_Mat_Mat_Mult(A, I, C, 1.0, 1.0);

ergibt C = A + C und lustigerweise ist letzteres trotz der  
zusätzlichen dritten Matrix anscheinend deutlich schneller als der  
operator+() weiter oben.

Gruß

Christian

Re: [Lapackpp-devel] lapackpp

[Christian S. <sti...@tu...>]

Dear Keith,

Am Donnerstag, 22. Oktober 2009 01:00 schrieben Sie:
> Mr. Stimming, we are attempting to use lapack++ in our application.  We are
> working in a multi-thread environment and are experiencing issues

Please define "experiencing issues".

> with our data results and we believe that thread issues in lapack++ may be
> the cause. To your knowledge, is lapack++ thread safe?  

Please define "thread safe". The data container classes (LaGenMatDouble etc) 
in lapack++ are just that - data container classes. Those are not protected 
against access from multiple threads concurrently for good reasons (way way 
too high runtime cost in the normal use case). If you planned to access any 
of the data classes from multiple threads concurrently, you must set up your 
multi-threading protection on top of the lapack++ classes.

Other than that, all computational functions are thread safe in the sense that 
as long as they work on different container class objects, calling the 
computational functions from multiple threads concurrently will not access 
any common (e.g. static) storage and hence is thread safe. In particular, any 
temporary workspace memory is allocated in each function call separately 
which means this is indeed thread safe.

Regards,

Christian Stimming

> Any response that you 
> can give would be helpful.  Thanks.
>
>
>
> Keith Blonquist
>
> Junior Engineer
>
> Lidar Pacific Corporation

[Lapackpp-devel] Newbie question: [SVD] Getting Singular Values?

[Ariady P. <ari...@ya...>]

hello all,
i'm having trouble getting Singular Values from a matrix. if anyone would give a simple sample code how to compute SVD it would be very appreciated.
thanks before!

btw, i need the Singular Values in String form (i'm working with Windows Form Application).


      New Email names for you! 
Get the Email name you've always wanted on the new @ymail and @rocketmail. 
Hurry before someone else does!
http://mail.promotions.yahoo.com/newdomains/aa/

Re: [Lapackpp-devel] lapackpp matrix type conversion

[Christian S. <sti...@tu...>]

Am Donnerstag, 30. Oktober 2008 00:46 schrieb Matthew Wilkins:
> How do I convert between Symmetric Double and General Double matrices and
> vice versa? For example, I have a matrix A defined in my class definition
> as LaSymmMatDouble. To compute this matrix, I have to multiply a non-square
> LaGenMatDouble matrix B with it's transpose. I would like to convert the
> LaGenMatDouble result and save it as a LaSymmMatDouble for use in further
> equations where I will take advantage of the BLAS symmetric multiplication
> routines. Is there an easy way to do this?
>
> int n = 5;
> LaGenMatDouble B(n,n),C(n,n);
> LaSymmMatDouble A(n,n);
>
> Blas_Mat_Trans_Mat_Mult(B,B,C);
>
> // How do I accomplish this step?
> A = C;

There is no function available in lapackpp to do this. Sorry.

I don't even know what lapack/blas (without the"pp") offer for this kind of 
conversion. Maybe there is a multiplication available whose output type is 
directly the symmetric matrix, as this is the only way to ensure a symmetric 
matrix really results.

> I tried to use the la::convert_mat<type>(A) function but it apparently does
> not contain the mapping between the symmetric and general matrix classes.

Right; those functions are only convert between the value_type of the matrix, 
i.e. between double, float, int, and long int.

Regards,

Christian

Re: [Lapackpp-devel] a problem on LaIndex class operator +

[Christian S. <sti...@tu...>]

Am Dienstag, 10. Juni 2008 10:11 schrieb 小小河/tp:
> hi stimming,
> i am a new user of lapack++2.5.2. i have trouble with the LaIndex class
> operator +. your code in laindex.h is that
> ---------
>       inline LaIndex& operator+(int i) {
>      start_+=i; end_+=i; return *this;}
> ---------
> the problem is that, when the operator + is used like this:
> -------------
> LaIndex i(0,1);
> LaGenMatDouble A(3,3);
> std::cout << A(i+1,i+1);
> -------------
> the third line will cause an error, because i has been changed to be
> (2:1:3) not expected (0:1:1).
>
> I think the following definition maybe better:
> ------------
> inline LaIndex operator+(int i);
> ------------

Absolutely. Thanks for pointing this out. I've committed this to SVN and it 
will be in the next release.

Christian

Re: [Lapackpp-devel] Dose lapackpp2.5.2 have functions for eigenvalue prolem Ax=kBx, with A, B are matrices?

[Christian S. <sti...@tu...>]

Questions about lapackpp to the mailing list 
lap...@li... , *not* to individual developers! 
Thanks.

You are looking for "General Eigenvalue" problem solving? No, lapackpp doesn't 
have this. Lapack (without pp) does have it, though, so it's a matter of 
writing a suitable wrapper function. Such a wrapper function would look 
similar to the existing LaEigSolve().

Christian

Am Dienstag, 10. Juni 2008 13:49 schrieb 小小河/tp:
> hi stimming,
>
> I am sorry to trouble again.
> Dose lapack++2.5.2 contain functions which can solve the eigenvalue
> problem: Ax=kBx, where A and B are matrices, k is eigenvalue.
>
> thank you very much!
>
> ------------------
> 中国科学院数学与系统科学研究院计算数学与科学工程计算研究所
> 北京市海淀区中关村东路55号,100190
> 何沧平
> He Cangping
>
> Academy of Mathematics and Systems Science,  Chinese  Academy of  Sciences.
> No.55 East Road, Zhongguancun, Beijing 100190,P.R. China

[Lapackpp-devel] [NewBie] LAPACK++ in Windows XP with DEV-C++

[imtnan q. <im...@gm...>]

Hello All,

I have downloaded and installed
lapackpp-2.5.1-setup.exe<http://downloads.sourceforge.net/lapackpp/lapackpp-2.5.1-setup.exe?modtime=1170171328&big_mirror=0>from
sourceforge.net.
I am working in Windows XP environement in Blood Shed Dev_C++. The problem ,
I am facing
is the basic configuration of the Dev-C++ environment for the use of
LAPACK++.

Second problem is regarding the complex matrices. I am unable to compile any
example for complex matrices.

Please If someone can help. I shall be gratefull,

Thanx in advance,

-- 
Imtnan-Ul-Haque QAZI

Doctorant Laboratoire du SIC,
Bât. SP2MI, Téléport 2,
Bvd Marie et Pierre Curie, BP 30179
86962, Futuroscope Chasseneuil Cedex France
http://www.sic.sp2mi.univ-poitiers.fr
Tel fixe : 00 33 (0)5 49 49 65 11
Cell : 00 33 (0)6 43 19 60 96
Fax : 00 33 (0)5 49 49 65 70

Re: [Lapackpp-devel] (Real) double QR factorization

[Christian S. <sti...@tu...>]

Quoting Brian White <bw...@ca...>:
> I have need for a real QR factorization.  i.e., the analog of
> LaGenQRFactComplex::generateQ_IP from matrix/src/gfqrc.cc.  I don't
> believe this functionality is redundant with what exists, since
> LaGenMatDouble and LaGenMatComplex have no inheritance relation.  (...)
>
> In any case, I have done this for an old version of lapackpp.  It is
> straightforward and amounts to creating gfqrd.cc from gfqrc.cc, in which
> LaGenMatComplex -> LaGenMatDouble and the complex Fortran routines
> (zgeqrf, zungqr, zunmqr) -> the real routines (dgeqrf, dorgqr, dormqr).
>
> Is there interest in adding this to the distribution?

Yes, if you already have the code, you are very welcome to contribute  
this to the project. If you're adding new functionality, and this  
functionality exists as separate lapack functions like dgeqrf etc.,  
you can just go ahead and add this to the up-to-date lapackpp.

>  Years ago, I was
> listed as a developer, but would rather not make such a substantial change
> behind Christian's back.  Please let me know if there is interest in this
> extension and, if so, how to proceed (e.g., provide patches to version
> 2.5.2, directly access cvs, etc.).

If you still have the sourceforce account, you can directly submit  
your code to SVN (no longer CVS). That would be the easiest way to  
contribute.

Regards,

Christian

[Lapackpp-devel] (Real) double QR factorization

[Brian W. <bw...@ca...>]

Hello,

I have need for a real QR factorization.  i.e., the analog of
LaGenQRFactComplex::generateQ_IP from matrix/src/gfqrc.cc.  I don't
believe this functionality is redundant with what exists, since
LaGenMatDouble and LaGenMatComplex have no inheritance relation.  Further,
I don't want to waste memory copying my real data to a complex matrix.
I suspect that the Fortran complex routines are not as efficient as the
real ones anyway; though I have nothing but intuition to back that up.

In any case, I have done this for an old version of lapackpp.  It is
straightforward and amounts to creating gfqrd.cc from gfqrc.cc, in which
LaGenMatComplex -> LaGenMatDouble and the complex Fortran routines
(zgeqrf, zungqr, zunmqr) -> the real routines (dgeqrf, dorgqr, dormqr).

Is there interest in adding this to the distribution?  Years ago, I was
listed as a developer, but would rather not make such a substantial change
behind Christian's back.  Please let me know if there is interest in this
extension and, if so, how to proceed (e.g., provide patches to version
2.5.2, directly access cvs, etc.).

Thank you,
Brian

Re: [Lapackpp-devel] Computing a single eigenvector

[Christian S. <sti...@tu...>]

Hi!

Am Donnerstag, 3. April 2008 20:14 schrieb Cody Planteen:
> I am using LAPACK++ with some very large symmetric matricies for which
> I need to compute eigenvectors.  However, I only need the eigenvector
> associated with the second smallest eigenvalue, rather than all
> eigenvectors of the matrix.  Are there any LAPACK++ functions to
> compute a single eigenvector? 

No.

> I have done some research on LAPACK 
> functions and the "expert" function SSYEVX allows for computation of a
> range of indicies, rather than all eigenvalues/eigenvectors.

If SSYEVX does what you want, you should probably use it directly in *your* 
source code, rather than trying to fit it somewhere into LAPACK++. I think 
this would be the easiest solution.

> If I 
> were to modify the LAPACK++ source to implement this, are there any
> naming conventions for the function or other considerations that I
> should be aware of?

There are almost no naming conventions. However, I doubt whether there is an 
easy possibility to design a C++ function that offers all functionality of 
SSYEVX through some meaningful arguments, hence I'd rather propose you should 
use SSYEVX yourself. But if you think it's easier for you to do this in 
LAPACK++, we're open for patches :-)

Christian

[Lapackpp-devel] Computing a single eigenvector

[Cody P. <pla...@gm...>]

Hello,

I am using LAPACK++ with some very large symmetric matricies for which
I need to compute eigenvectors.  However, I only need the eigenvector
associated with the second smallest eigenvalue, rather than all
eigenvectors of the matrix.  Are there any LAPACK++ functions to
compute a single eigenvector? I have done some research on LAPACK
functions and the "expert" function SSYEVX allows for computation of a
range of indicies, rather than all eigenvalues/eigenvectors.  If I
were to modify the LAPACK++ source to implement this, are there any
naming conventions for the function or other considerations that I
should be aware of?

Thanks,

Cody

Re: [Lapackpp-devel] [Newbie] Support for Windows

[Christian S. <sti...@tu...>]

Am Freitag, 8. Februar 2008 21:59 schrieb Stu Schaffner:
> I have an MS Windows Vista system with MS Visual Studio 2005, and I wish to
> use LAPACK++.  I have no FORTRAN compiler.  I downloaded lapackpp-2.5.2 as
> instructed.  The README suggests that I should run a .exe file to install
> the packages blas and lapack.  The .exe file does not seem to be in the
> distribution, although I saw a posting that indicated that it was available
> in earlier versions.  However, if I understand this system correctly, the
> .exe file will simply install DLLs for blas and lapack.  Since Vista
> doesn't store DLLs in the same directories, it would seem easier to just
> copy the DLLs in manually.  Is this the best approach?  If so, where can I
> get the DLLs?

Here: http://lapackpp.cvs.sourceforge.net/lapackpp/lapackpp/contrib/

I'm tired of pointing people to somewhere else, so I just uploaded the DLLs to 
CVS.

Christian

[Lapackpp-devel] [Newbie] Support for Windows

[Stu S. <st...@rc...>]

I have an MS Windows Vista system with MS Visual Studio 2005, and I wish to
use LAPACK++.  I have no FORTRAN compiler.  I downloaded lapackpp-2.5.2 as
instructed.  The README suggests that I should run a .exe file to install
the packages blas and lapack.  The .exe file does not seem to be in the
distribution, although I saw a posting that indicated that it was available
in earlier versions.  However, if I understand this system correctly, the
.exe file will simply install DLLs for blas and lapack.  Since Vista doesn't
store DLLs in the same directories, it would seem easier to just copy the
DLLs in manually.  Is this the best approach?  If so, where can I get the
DLLs?

 

Stu Schaffner

Re: [Lapackpp-devel] thread safety of lapack++

[Christian S. <sti...@tu...>]

Am Samstag, 1. Dezember 2007 07:49 schrieb Matt Hansen:
> I've noticed that some of the functions in lapack++ do not appear to
> be thread safe, which is inconvenient when programming for new
> multi-core processors, although not impossible to work around. I was
> wondering if anyone has a comprehensive list of which functions are
> not thread safe.

I'm curious what you mean by "not thread-safe". 

Are you talking about static class variables? Those exist, but are limited to 
debugging output and formatting of that output - they don't affect any normal 
computations.

Or are you talking about locking access to the object content? Then I have to 
disappoint you - the access methods of all matrix classes do *not* include 
any form of thread synchronization (i.e. an access mutex). This is 
intentional and it will stay this way, because introducing thread 
synchronization at the level of the matrix classes is the wrong level of 
dealing with this. Imagine a "thread-safe matrix class": It would have to 
contain a mutex, which on each operator()(int,int) access will get locked, 
the element will be returned, and the mutex will be unlocked. Now imagine 
someone using the operator()(int,int) in a for-loop. You will have O(n^2) 
mutex locking/unlocking in that loop, which is just insane. (Similar 
arguments explain why std::vector<> is not "thread-safe" and never should 
be.)

No, dealing with thread synchronization has to be done on a higher layer, i.e. 
outside of Lapackpp. In that sense, all Lapackpp's classes and functions are 
indeed not "thread-safe".

Christian

> I'm currently using the following functions and data structures in my code:
> LaGenMatDouble
> LaGenMatComplex
> LaVectorDouble
> LaVectorComplex
> LaVectorLongInt
>
> LaEigSolve
> Blas_Mat_Mat_Mult
> LuFactorizeIP
> LaLUInverseIP
>
> Thank you in advance for any insight you can offer,
>
> Matt Hansen
> Electrical Engineering and Physics at MTU

[Lapackpp-devel] thread safety of lapack++

[Matt H. <con...@gm...>]

I've noticed that some of the functions in lapack++ do not appear to
be thread safe, which is inconvenient when programming for new
multi-core processors, although not impossible to work around. I was
wondering if anyone has a comprehensive list of which functions are
not thread safe.

I'm currently using the following functions and data structures in my code:
LaGenMatDouble
LaGenMatComplex
LaVectorDouble
LaVectorComplex
LaVectorLongInt

LaEigSolve
Blas_Mat_Mat_Mult
LuFactorizeIP
LaLUInverseIP

Thank you in advance for any insight you can offer,

Matt Hansen
Electrical Engineering and Physics at MTU

Re: [Lapackpp-devel] question about lapack++ 2.5.2

[Christian S. <sti...@tu...>]

Questions about lapackpp *always* to the mailing list lapackpp-devel, =20
*never* only to individual developers!!!!!!!!

Quoting Fernando Pizarro <piz...@go...>:
> I am trying to use lapack++ 2.5.2 with (MSVC) following the instruction of
> you website:
>
> For compiling on Windows with the Microsoft Visual Studio C++ (MSVC)
> compiler:
>
> 1. Install the .exe setup (...)
>
> but I can find the .exe file, isn't inside of the all platform file on
> http://www.sourceforge.net/projects/lapackpp,

Use the earlier version of the exe file, then compile the last version =20
from the source code.

Christian



>
> Can you provide of more information, I really want to use this library
> because has all the functionality I need. I will notify  you if I am able =
to
> use the library on windows vista (where I am currently working).
>
> Thanks in advance for you help and congratulations for this excellent job
> that you and other are making...
>
> greetings,
>
>   Fernando.
>
> PD: if you are not able to help me, can you notify me please... thanks
> again.
>

Re: [Lapackpp-devel] how to use clapack in lapackpp?

[Christian S. <sti...@tu...>]

Quoting Wenjian Zhang <we...@us...>:
> Hi,
>
> Please hlep me on this issue. I am working on a project at
> windows platform. How can i compile the lapackpp to use
> clapack (not lapack in fortran)?

You can't.

Christian

Re: [Lapackpp-devel] Trouble compiling lapack++ in MSVC

[Christian S. <sti...@tu...>]

Am Mittwoch, 29. August 2007 11:13 schrieb k.s...@fy...:
> I'm having some trouble compiling lapack++ 2.5.2 using MSVC (ver.
> 7.1.6030).
>
> The debug build results in lots of these:
> include\laexcp.h(74) : warning C4275: non dll-interface class
> 'std::runtime_error' used as base for dll-interface class 'LaException'
>         C:\Program Files\Microsoft Visual Studio .NET
> 2003\Vc7\include\stdexcept(136) : see declaration of
> 'std::runtime_error'
>         include\laexcp.h(73) : see declaration of 'LaException'

Those are normal and can be ignored.

> And the release build just says:
> lapackpp : error PRJ0003 : Error spawning ''.
>
> I've put up the entire buildlogs here:
> http://folk.uio.no/knutsj/err/BuildLog_debug.htm
> http://folk.uio.no/knutsj/err/BuildLog_release.htm
>
>
> The debug build does produce .lib and .dll files, but when I try to build
> tgd i get a couple dozen errors like this:
> tgd.obj : error LNK2019: unresolved external symbol "__declspec(dllimport)
> public: class LaGenMatDouble & __thiscall
> LaGenMatDouble::operator=(double)" (__imp_??4LaGenMatDouble@@QAEAAV0@N@Z)
> referenced in function "void __cdecl la::ones<class LaGenMatDouble>(class
> LaGenMatDouble &,int,int)"
> (??$ones@VLaGenMatDouble@@@la@@YAXAAVLaGenMatDouble@@HH@Z)

Those I saw just recently and I don't have a good idea why they show up. The 
function in question (LaGenMatDouble::operator=) exists and its 
implementation exists as well. If you use the dependency walker 
(depends.exe), you will find this symbol in the lapackpp32.dll as well. On 
Linux there is no such problem, so unfortunately I don't know how to fix this 
at this point in time.

Christian

[Lapackpp-devel] Trouble compiling lapack++ in MSVC

[<k.s...@fy...>]

I'm having some trouble compiling lapack++ 2.5.2 using MSVC (ver. 7.1.6030).

The debug build results in lots of these:
include\laexcp.h(74) : warning C4275: non dll-interface class
'std::runtime_error' used as base for dll-interface class 'LaException'
        C:\Program Files\Microsoft Visual Studio .NET
2003\Vc7\include\stdexcept(136) : see declaration of
'std::runtime_error'
        include\laexcp.h(73) : see declaration of 'LaException'


And the release build just says:
lapackpp : error PRJ0003 : Error spawning ''.

I've put up the entire buildlogs here:
http://folk.uio.no/knutsj/err/BuildLog_debug.htm
http://folk.uio.no/knutsj/err/BuildLog_release.htm


The debug build does produce .lib and .dll files, but when I try to build
tgd i get a couple dozen errors like this:
tgd.obj : error LNK2019: unresolved external symbol "__declspec(dllimport)
public: class LaGenMatDouble & __thiscall
LaGenMatDouble::operator=(double)" (__imp_??4LaGenMatDouble@@QAEAAV0@N@Z)
referenced in function "void __cdecl la::ones<class LaGenMatDouble>(class
LaGenMatDouble &,int,int)"
(??$ones@VLaGenMatDouble@@@la@@YAXAAVLaGenMatDouble@@HH@Z)


Any help is appreciated.

Knut S J