209 lines
9.2 KiB
Plaintext
209 lines
9.2 KiB
Plaintext
-------------------------------------------------------------------------------
|
|
C Interface to LAPACK
|
|
README
|
|
-------------------------------------------------------------------------------
|
|
Introduction
|
|
-------------------------------------------------------------------------------
|
|
|
|
This library is a part of reference implementation for the C interface to
|
|
LAPACK project according to the specifications described at the forum for
|
|
the Intel(R) Math Kernel Library (Intel(R) MKL):
|
|
http://software.intel.com/en-us/forums/showthread.php?t=61234
|
|
|
|
This implementation provides a native C interface to LAPACK routines available
|
|
at www.netlib.org/lapack to facilitate usage of LAPACK functionality
|
|
for C programmers.
|
|
This implementation introduces:
|
|
- row-major and column-major matrix layout controlled by the first function
|
|
parameter;
|
|
- an implementation with working arrays (middle-level interface) as well as
|
|
without working arrays (high-level interface);
|
|
- input scalars passed by value;
|
|
- error code as a return value instead of the INFO parameter.
|
|
|
|
This implementation supports both the ILP64 and LP64 programming models,
|
|
and different complex type styles: structure, C99.
|
|
|
|
This implementation includes interfaces for the LAPACK-3.2.1 Driver and
|
|
Computational routines only.
|
|
|
|
-------------------------------------------------------------------------------
|
|
Product Directories
|
|
-------------------------------------------------------------------------------
|
|
|
|
The installation directory of this package has the following structure:
|
|
|
|
src - C interface source files
|
|
utils - C interface auxiliary files
|
|
include - header files for C interface
|
|
|
|
-------------------------------------------------------------------------------
|
|
Installation
|
|
-------------------------------------------------------------------------------
|
|
|
|
The reference code for the C interface to LAPACK is built similarly to the
|
|
Basic Linear Algebra Subprograms (BLAS) and LAPACK. The build system produces
|
|
a static binary lapacke.a.
|
|
|
|
You need to provide a make.inc file in the top directory that defines the
|
|
compiler, compiler flags, names for binaries to be created/linked to. You may
|
|
choose the appropriate LP64/ILP64 model, convenient complex type style,
|
|
LAPACKE name pattern, and/or redefine system malloc/free in make.inc. Several
|
|
examples of make.inc are provided.
|
|
|
|
After setting up the make.inc, you can build C interface to LAPACK by typing
|
|
|
|
make lapacke
|
|
|
|
-------------------------------------------------------------------------------
|
|
Handling Complex Types
|
|
-------------------------------------------------------------------------------
|
|
|
|
The interface uses complex types lapack_complex_float/lapack_complex_double.
|
|
You have several options to define them:
|
|
|
|
1) C99 complex types (default):
|
|
|
|
#define lapack_complex_float float _Complex
|
|
#define lapack_complex_double double _Complex
|
|
|
|
2) C structure option (set by enabling in the configuration file):
|
|
-DHAVE_LAPACK_CONFIG_H -DLAPACK_COMPLEX_STRUCTURE
|
|
|
|
typedef struct { float real, imag; } _lapack_complex_float;
|
|
typedef struct { double real, imag; } _lapack_complex_double;
|
|
#define lapack_complex_float _lapack_complex_float
|
|
#define lapack_complex_double _lapack_complex_double
|
|
|
|
3) C++ complex types (set by enabling in the configuration file):
|
|
-DHAVE_LAPACK_CONFIG_H -DLAPACK_COMPLEX_CPP
|
|
|
|
#define lapack_complex_float std::complex<float>
|
|
#define lapack_complex_double std::complex<double>
|
|
|
|
You have to compile the interface with C++ compiler with C++ types.
|
|
|
|
4) Custom complex types:
|
|
-DLAPACK_COMPLEX_CUSTOM
|
|
|
|
To use custom complex types, you need to:
|
|
- Define lapack_complex_float/lapack_complex_double types on your own.
|
|
- Optionally define lapack_make_complex_float/lapack_make_complex_double_real
|
|
functions if you want to build the testing suite supplied. Use these
|
|
functions for the testing system. Their purpose is to make a complex value of
|
|
a real part re, imaginary part im. The prototypes are as follows:
|
|
|
|
lapack_complex_float lapack_make_complex_float( float re, float im );
|
|
lapack_complex_double lapack_make_complex_double( double re, double im );
|
|
|
|
-------------------------------------------------------------------------------
|
|
Choosing ILP64 Data Model
|
|
-------------------------------------------------------------------------------
|
|
To choose ILP64 data model (set by enabling in the configuration file), use the
|
|
following options:
|
|
|
|
-DHAVE_LAPACK_CONFIG_H -DLAPACK_ILP64
|
|
|
|
-------------------------------------------------------------------------------
|
|
Using Predicate Functions
|
|
-------------------------------------------------------------------------------
|
|
|
|
The functions
|
|
|
|
lapacke_?gees/lapacke_?gees_work
|
|
lapacke_?geesx/lapacke_?geesx_work
|
|
lapacke_?geev/lapacke_?geev_work
|
|
lapacke_?geevx/lapacke_?geevx_work
|
|
|
|
require the pointer to a predicate function as an argument of a predefined type
|
|
such as:
|
|
|
|
typedef lapack_logical (*LAPACK_S_SELECT2) ( const float*, const float* );
|
|
|
|
The purpose and format of these predicate functions are described in the LAPACK
|
|
documentation. This interface passes the pointer to the corresponding LAPACK
|
|
routine as it is.
|
|
|
|
Be cautious with return values of the logical type if you link against LAPACK
|
|
compiled with Fortran compiler. Whereas all non-zero values are treated as TRUE
|
|
generally, some Fortran compilers may rely on a certain TRUE value, so you will
|
|
have to use the same TRUE value in the predicate function to be consistent with
|
|
LAPACK implementation.
|
|
|
|
-------------------------------------------------------------------------------
|
|
Implementation Details
|
|
-------------------------------------------------------------------------------
|
|
|
|
The current C interface implementation consists of wrappers to LAPACK routines.
|
|
The row-major matrices are transposed on entry to and on exit from the LAPACK
|
|
routine, if needed. Top-level interfaces additionally allocate/deallocate
|
|
working space on entry to and on exit from the LAPACK routine.
|
|
|
|
Because of possible additional transpositions, a routine called with
|
|
this interface may require more memory space and run slower than the
|
|
corresponding LAPACK routine.
|
|
|
|
-------------------------------------------------------------------------------
|
|
Disclaimer and Legal Information
|
|
-------------------------------------------------------------------------------
|
|
|
|
INFORMATION IN THIS DOCUMENT IS PROVIDED IN CONNECTION WITH INTEL(R)
|
|
PRODUCTS. NO LICENSE, EXPRESS OR IMPLIED, BY ESTOPPEL OR OTHERWISE, TO
|
|
ANY INTELLECTUAL PROPERTY RIGHTS IS GRANTED BY THIS DOCUMENT. EXCEPT
|
|
AS PROVIDED IN INTEL'S TERMS AND CONDITIONS OF SALE FOR SUCH PRODUCTS,
|
|
INTEL ASSUMES NO LIABILITY WHATSOEVER, AND INTEL DISCLAIMS ANY EXPRESS
|
|
OR IMPLIED WARRANTY, RELATING TO SALE AND/OR USE OF INTEL PRODUCTS
|
|
INCLUDING LIABILITY OR WARRANTIES RELATING TO FITNESS FOR A PARTICULAR
|
|
PURPOSE, MERCHANTABILITY, OR INFRINGEMENT OF ANY PATENT, COPYRIGHT OR
|
|
OTHER INTELLECTUAL PROPERTY RIGHT. UNLESS OTHERWISE AGREED IN WRITING
|
|
BY INTEL, THE INTEL PRODUCTS ARE NOT DESIGNED NOR INTENDED FOR ANY
|
|
APPLICATION IN WHICH THE FAILURE OF THE INTEL PRODUCT COULD CREATE A
|
|
SITUATION WHERE PERSONAL INJURY OR DEATH MAY OCCUR.
|
|
|
|
Intel may make changes to specifications and product descriptions at
|
|
any time, without notice. Designers must not rely on the absence or
|
|
characteristics of any features or instructions marked "reserved" or
|
|
"undefined." Intel reserves these for future definition and shall have
|
|
no responsibility whatsoever for conflicts or incompatibilities
|
|
arising from future changes to them. The information here is subject
|
|
to change without notice. Do not finalize a design with this
|
|
information.
|
|
|
|
The products described in this document may contain design defects or
|
|
errors known as errata which may cause the product to deviate from
|
|
published specifications. Current characterized errata are available
|
|
on request.
|
|
|
|
Contact your local Intel sales office or your distributor to obtain
|
|
the latest specifications and before placing your product order.
|
|
Copies of documents which have an order number and are referenced in
|
|
this document, or other Intel literature, may be obtained by calling
|
|
1-800-548-4725, or go to http://www.intel.com/design/literature.htm
|
|
|
|
Intel processor numbers are not a measure of performance. Processor
|
|
numbers differentiate features within each processor family, not
|
|
across different processor families. See
|
|
http://www.intel.com/products/processor_number for details.
|
|
|
|
This document contains information on products in the design phase of
|
|
development.
|
|
|
|
BunnyPeople, Celeron, Celeron Inside, Centrino, Centrino Atom,
|
|
Centrino Atom Inside, Centrino Inside, Centrino logo, Core Inside,
|
|
FlashFile, i960, InstantIP, Intel, Intel logo, Intel386, Intel486,
|
|
IntelDX2, IntelDX4, IntelSX2, Intel Atom, Intel Atom Inside, Intel
|
|
Core, Intel Inside, Intel Inside logo, Intel. Leap ahead., Intel. Leap
|
|
ahead. logo, Intel NetBurst, Intel NetMerge, Intel NetStructure,
|
|
Intel SingleDriver, Intel SpeedStep, Intel StrataFlash, Intel Viiv,
|
|
Intel vPro, XScale, IPLink, Itanium, Itanium Inside, MCS, MMX, Oplus,
|
|
OverDrive, Intel PDCharm, Pentium, Pentium Inside, skoool, Sound Mark,
|
|
The Journey Inside, VTune, Xeon, and Xeon Inside are trademarks of
|
|
Intel Corporation in the U.S. and other countries.
|
|
|
|
|
|
* Other names and brands may be claimed as the property of others.
|
|
|
|
Copyright (C) 2011, Intel Corporation. All rights reserved.
|
|
|
|
|