Difference between revisions of "GPU610/DPS915 Student Resources"

From CDOT Wiki
Jump to: navigation, search
(BLAS Documentation)
Line 4: Line 4:
  
 
==BLAS Documentation==
 
==BLAS Documentation==
 
+
See the [[GPU610/DPS915_BLAS_Documentation | BLAS Documentation Page]]
Note: This information applies to the '''gsl_cblas''' library 
 
 
 
There are 2 main functions to use
 
 
 
===segmv===
 
* compute the matrix-vector product for a general matrix
 
 
 
void '''cblas_sgemv''' (''const enum CBLAS_ORDER '''order''', const enum CBLAS_TRANSPOSE '''TransA''', const int '''M''', const int '''N''', const float '''alpha''', const float * '''A''', const int '''lda''', const float * '''x''', const int '''incx''', const float '''beta''', float * '''y''', const int '''incy''''')[http://www.gnu.org/software/gsl/manual/html_node/Level-2-CBLAS-Functions.html 1] [http://www.prism.gatech.edu/~ndantam3/cblas-doc/doc/html/cblas_8h.html#23ac27150577c29a7ad4ddb427f255f7 2] [http://publib.boulder.ibm.com/infocenter/comphelp/v8v101/index.jsp?topic=%2Fcom.ibm.xlcpp8a.doc%2Fproguide%2Fref%2Fblaslib.htm 3]
 
 
 
 
 
'''''order'''''
 
:Whether matrices are row major order (C-Style) for column major order (Fortran-style). One of enum CblasRowMajor or CblasColMajor.
 
 
 
'''''TransA'''''
 
:Whether to transpose matrix A. One of enum CblasNoTrans, CBlasTrans.
 
 
 
'''''M'''''
 
:represents:
 
::the number of rows in input matrix a
 
:The number of rows must be greater than or equal to zero, and less than the leading dimension of the matrix a (specified in lda)
 
 
 
'''''N'''''
 
:represents:
 
::the number of columns in input matrix a
 
:The number of columns must be greater than or equal to zero.
 
 
 
'''''alpha'''''
 
:is the scaling constant for matrix a
 
 
 
'''''A'''''
 
:is the input matrix of float (for sgemv) or double (for dgemv) values
 
 
 
'''''lda'''''
 
:is the leading dimension of the array specified by a. The leading dimension must be greater than zero. The leading dimension must be greater than or equal to 1 and greater than or equal to the value specified in m.
 
 
 
'''''x'''''
 
:is the input vector of float (for sgemv) or double (for dgemv) values.
 
 
 
'''''incx'''''
 
:is the stride for vector x. It can have any value.
 
 
 
'''''beta'''''
 
:is the scaling constant for vector y
 
 
 
'''''y'''''
 
:is the output vector of float (for sgemv) or double (for dgemv) values.
 
 
 
'''''incy'''''
 
:is the stride for vector y. It must not be zero.
 
 
 
===segmm===
 
*  perform combined matrix multiplication and addition for general matrices
 
 
 
void '''cblas_sgemm''' (''const enum CBLAS_ORDER '''Order''', const enum CBLAS_TRANSPOSE '''TransA''', const enum CBLAS_TRANSPOSE '''TransB''', const int '''M''', const int '''N''', const int '''K''', const float '''alpha''', const float * '''A''', const int '''lda''', const float * '''B''', const int '''ldb''', const float '''beta''', float * '''C''', const int '''ldc''''')[http://www.gnu.org/software/gsl/manual/html_node/Level-3-CBLAS-Functions.html 1] [http://www.prism.gatech.edu/~ndantam3/cblas-doc/doc/html/cblas_8h.html#7d42dfcb6073c56391fee28494809cc5 2] [http://publib.boulder.ibm.com/infocenter/comphelp/v8v101/index.jsp?topic=%2Fcom.ibm.xlcpp8a.doc%2Fproguide%2Fref%2Fblaslib.htm 3]
 
 
 
 
 
 
 
'''''Order'''''
 
:Whether matrices are row major order (C-Style) for column major order (Fortran-style). One of enum CblasRowMajor or CblasColMajor.
 
 
 
'''''TransA'''''
 
:Whether to transpose matrix A. One of enum CblasNoTrans, CBlasTrans, CBlasConjTrans
 
 
 
'''''TransB'''''
 
:Whether to transpose matrix B. One of enum CblasNoTrans, CBlasTrans, CBlasConjTrans.
 
 
 
'''''M'''''
 
: is the number of  Rows in matrices A and C
 
:M must be greater than or equal to zero.
 
 
 
'''''N'''''
 
:is the number of Columns in Matrices B and C
 
 
 
'''''K'''''
 
:is the number of Columns in matrix A and Rows in matrix B
 
 
 
'''''alpha'''''
 
:is the scaling constant for matrix a
 
 
 
'''''A'''''
 
:is the input matrix a of float (for sgemm).
 
 
 
'''''lda'''''
 
:is the leading dimension of the array specified by a. The leading dimension must be greater than zero. If transa is specified as 'N' or 'n', the leading dimension must be greater than or equal to 1. If transa is specified as 'T' or 't', the leading dimension must be greater than or equal to the value specified in m.
 
 
 
'''''B'''''
 
:is the input matrix b of float (for sgemm).
 
 
 
'''''ldb'''''
 
:is the leading dimension of the array specified by b. The leading dimension must be greater than zero. If transb is specified as 'N' or 'n', the leading dimension must be greater than or equal to the value specified in m. If transa is specified as 'T' or 't', the leading dimension must be greater than or equal to the value specified in n.
 
 
 
'''''beta'''''
 
:is the scaling constant for matrix c
 
 
 
'''''C'''''
 
:is the output matrix c of float (for sgemm) or double (for dgemm) values.
 
 
 
'''''ldc'''''
 
:is the leading dimension of the array specified by c. The leading dimension must be greater than zero. If transb is specified as 'N' or 'n', the leading dimension must be greater than or equal to 0 and greater than or equal to the value specified in l.
 
 
 
https://scs.senecac.on.ca/~gpu610/pages/content/lines.html
 
  
 
==Getting Started on Mac==
 
==Getting Started on Mac==

Revision as of 16:54, 20 September 2012


GPU610/DPS915 | Student List | Group and Project Index | Student Resources | Glossary

Student Resources

The purpose of this page is to share useful information that can help groups with their CUDA projects.

BLAS Documentation

See the BLAS Documentation Page

Getting Started on Mac

http://developer.download.nvidia.com/compute/DevZone/docs/html/C/doc/CUDA_Getting_Started_Mac.pdf

http://developer.nvidia.com/cuda/cuda-downloads

Troubleshooting

Problem with CUDA driver version 5.0.24 on MacBook Pro 2012 Fix