clBLAS  2.11
 All Functions Typedefs Enumerations Enumerator Groups Pages
Functions
TRMM - Triangular matrix-matrix multiplication

Functions

clblasStatus clblasStrmm (clblasOrder order, clblasSide side, clblasUplo uplo, clblasTranspose transA, clblasDiag diag, size_t M, size_t N, cl_float alpha, const cl_mem A, size_t offA, size_t lda, cl_mem B, size_t offB, size_t ldb, cl_uint numCommandQueues, cl_command_queue *commandQueues, cl_uint numEventsInWaitList, const cl_event *eventWaitList, cl_event *events)
 Multiplying a matrix by a triangular matrix with float elements. Extended version. More...
 
clblasStatus clblasDtrmm (clblasOrder order, clblasSide side, clblasUplo uplo, clblasTranspose transA, clblasDiag diag, size_t M, size_t N, cl_double alpha, const cl_mem A, size_t offA, size_t lda, cl_mem B, size_t offB, size_t ldb, cl_uint numCommandQueues, cl_command_queue *commandQueues, cl_uint numEventsInWaitList, const cl_event *eventWaitList, cl_event *events)
 Multiplying a matrix by a triangular matrix with double elements. Extended version. More...
 
clblasStatus clblasCtrmm (clblasOrder order, clblasSide side, clblasUplo uplo, clblasTranspose transA, clblasDiag diag, size_t M, size_t N, FloatComplex alpha, const cl_mem A, size_t offA, size_t lda, cl_mem B, size_t offB, size_t ldb, cl_uint numCommandQueues, cl_command_queue *commandQueues, cl_uint numEventsInWaitList, const cl_event *eventWaitList, cl_event *events)
 Multiplying a matrix by a triangular matrix with float complex elements. Extended version. More...
 
clblasStatus clblasZtrmm (clblasOrder order, clblasSide side, clblasUplo uplo, clblasTranspose transA, clblasDiag diag, size_t M, size_t N, DoubleComplex alpha, const cl_mem A, size_t offA, size_t lda, cl_mem B, size_t offB, size_t ldb, cl_uint numCommandQueues, cl_command_queue *commandQueues, cl_uint numEventsInWaitList, const cl_event *eventWaitList, cl_event *events)
 Multiplying a matrix by a triangular matrix with double complex elements. Extended version. More...
 

Detailed Description

Function Documentation

clblasStatus clblasCtrmm ( clblasOrder  order,
clblasSide  side,
clblasUplo  uplo,
clblasTranspose  transA,
clblasDiag  diag,
size_t  M,
size_t  N,
FloatComplex  alpha,
const cl_mem  A,
size_t  offA,
size_t  lda,
cl_mem  B,
size_t  offB,
size_t  ldb,
cl_uint  numCommandQueues,
cl_command_queue *  commandQueues,
cl_uint  numEventsInWaitList,
const cl_event *  eventWaitList,
cl_event *  events 
)

Multiplying a matrix by a triangular matrix with float complex elements. Extended version.

Matrix-triangular matrix products:

  • \( B \leftarrow \alpha A B \)
  • \( B \leftarrow \alpha A^T B \)
  • \( B \leftarrow \alpha B A \)
  • \( B \leftarrow \alpha B A^T \)

where T is an upper or lower triangular matrix.

Parameters
[in]orderRow/column order.
[in]sideThe side of triangular matrix.
[in]uploThe triangle in matrix being referenced.
[in]transAHow matrix A is to be transposed.
[in]diagSpecify whether matrix is unit triangular.
[in]MNumber of rows in matrix B.
[in]NNumber of columns in matrix B.
[in]alphaThe factor of matrix A.
[in]offAOffset of the first element of the matrix A in the buffer object. Counted in elements.
[in]ABuffer object storing matrix A.
[in]ldaLeading dimension of matrix A. For detailed description, see clblasStrmm().
[out]BBuffer object storing matrix B.
[in]offBOffset of the first element of the matrix B in the buffer object. Counted in elements.
[in]ldbLeading dimension of matrix B. For detailed description, see clblasStrmm().
[in]numCommandQueuesNumber of OpenCL command queues in which the task is to be performed.
[in]commandQueuesOpenCL command queues.
[in]numEventsInWaitListNumber of events in the event wait list.
[in]eventWaitListEvent wait list.
[in]eventsEvent objects per each command queue that identify a particular kernel execution instance.
Returns
  • clblasSuccess on success;
  • clblasInvalidValue if either offA or offB exceeds the size of the respective buffer object;
  • the same error codes as clblasStrmm() otherwise.
clblasStatus clblasDtrmm ( clblasOrder  order,
clblasSide  side,
clblasUplo  uplo,
clblasTranspose  transA,
clblasDiag  diag,
size_t  M,
size_t  N,
cl_double  alpha,
const cl_mem  A,
size_t  offA,
size_t  lda,
cl_mem  B,
size_t  offB,
size_t  ldb,
cl_uint  numCommandQueues,
cl_command_queue *  commandQueues,
cl_uint  numEventsInWaitList,
const cl_event *  eventWaitList,
cl_event *  events 
)

Multiplying a matrix by a triangular matrix with double elements. Extended version.

Matrix-triangular matrix products:

  • \( B \leftarrow \alpha A B \)
  • \( B \leftarrow \alpha A^T B \)
  • \( B \leftarrow \alpha B A \)
  • \( B \leftarrow \alpha B A^T \)

where T is an upper or lower triangular matrix.

Parameters
[in]orderRow/column order.
[in]sideThe side of triangular matrix.
[in]uploThe triangle in matrix being referenced.
[in]transAHow matrix A is to be transposed.
[in]diagSpecify whether matrix is unit triangular.
[in]MNumber of rows in matrix B.
[in]NNumber of columns in matrix B.
[in]alphaThe factor of matrix A.
[in]ABuffer object storing matrix A.
[in]offAOffset of the first element of the matrix A in the buffer object. Counted in elements.
[in]ldaLeading dimension of matrix A. For detailed description, see clblasStrmm().
[out]BBuffer object storing matrix B.
[in]offBOffset of the first element of the matrix B in the buffer object. Counted in elements.
[in]ldbLeading dimension of matrix B. For detailed description, see clblasStrmm().
[in]numCommandQueuesNumber of OpenCL command queues in which the task is to be performed.
[in]commandQueuesOpenCL command queues.
[in]numEventsInWaitListNumber of events in the event wait list.
[in]eventWaitListEvent wait list.
[in]eventsEvent objects per each command queue that identify a particular kernel execution instance.
Returns
  • clblasSuccess on success;
  • clblasInvalidDevice if a target device does not support floating point arithmetic with double precision;
  • clblasInvalidValue if either offA or offB exceeds the size of the respective buffer object;
  • the same error codes as the clblasStrmm() function otherwise.
clblasStatus clblasStrmm ( clblasOrder  order,
clblasSide  side,
clblasUplo  uplo,
clblasTranspose  transA,
clblasDiag  diag,
size_t  M,
size_t  N,
cl_float  alpha,
const cl_mem  A,
size_t  offA,
size_t  lda,
cl_mem  B,
size_t  offB,
size_t  ldb,
cl_uint  numCommandQueues,
cl_command_queue *  commandQueues,
cl_uint  numEventsInWaitList,
const cl_event *  eventWaitList,
cl_event *  events 
)

Multiplying a matrix by a triangular matrix with float elements. Extended version.

Matrix-triangular matrix products:

  • \( B \leftarrow \alpha A B \)
  • \( B \leftarrow \alpha A^T B \)
  • \( B \leftarrow \alpha B A \)
  • \( B \leftarrow \alpha B A^T \)

where T is an upper or lower triangular matrix.

Parameters
[in]orderRow/column order.
[in]sideThe side of triangular matrix.
[in]uploThe triangle in matrix being referenced.
[in]transAHow matrix A is to be transposed.
[in]diagSpecify whether matrix is unit triangular.
[in]MNumber of rows in matrix B.
[in]NNumber of columns in matrix B.
[in]alphaThe factor of matrix A.
[in]ABuffer object storing matrix A.
[in]offAOffset of the first element of the matrix A in the buffer object. Counted in elements.
[in]ldaLeading dimension of matrix A. It cannot be less than M when the side parameter is set to clblasLeft,
or less than N when it is set to clblasRight.
[out]BBuffer object storing matrix B.
[in]offBOffset of the first element of the matrix B in the buffer object. Counted in elements.
[in]ldbLeading dimension of matrix B. It cannot be less than N when the order parameter is set to clblasRowMajor,
or not less than M when it is set to clblasColumnMajor.
[in]numCommandQueuesNumber of OpenCL command queues in which the task is to be performed.
[in]commandQueuesOpenCL command queues.
[in]numEventsInWaitListNumber of events in the event wait list.
[in]eventWaitListEvent wait list.
[in]eventsEvent objects per each command queue that identify a particular kernel execution instance.
Returns
  • clblasSuccess on success;
  • clblasInvalidValue if either offA or offB exceeds the size of the respective buffer object;
  • the same error codes as clblasStrmm() otherwise.
Examples:
example_strmm.c.
clblasStatus clblasZtrmm ( clblasOrder  order,
clblasSide  side,
clblasUplo  uplo,
clblasTranspose  transA,
clblasDiag  diag,
size_t  M,
size_t  N,
DoubleComplex  alpha,
const cl_mem  A,
size_t  offA,
size_t  lda,
cl_mem  B,
size_t  offB,
size_t  ldb,
cl_uint  numCommandQueues,
cl_command_queue *  commandQueues,
cl_uint  numEventsInWaitList,
const cl_event *  eventWaitList,
cl_event *  events 
)

Multiplying a matrix by a triangular matrix with double complex elements. Extended version.

Matrix-triangular matrix products:

  • \( B \leftarrow \alpha A B \)
  • \( B \leftarrow \alpha A^T B \)
  • \( B \leftarrow \alpha B A \)
  • \( B \leftarrow \alpha B A^T \)

where T is an upper or lower triangular matrix.

Parameters
[in]orderRow/column order.
[in]sideThe side of triangular matrix.
[in]uploThe triangle in matrix being referenced.
[in]transAHow matrix A is to be transposed.
[in]diagSpecify whether matrix is unit triangular.
[in]MNumber of rows in matrix B.
[in]NNumber of columns in matrix B.
[in]alphaThe factor of matrix A.
[in]ABuffer object storing matrix A.
[in]offAOffset of the first element of the matrix A in the buffer object. Counted in elements.
[in]ldaLeading dimension of matrix A. For detailed description, see clblasStrmm().
[out]BBuffer object storing matrix B.
[in]offBOffset of the first element of the matrix B in the buffer object. Counted in elements.
[in]ldbLeading dimension of matrix B. For detailed description, see clblasStrmm().
[in]numCommandQueuesNumber of OpenCL command queues in which the task is to be performed.
[in]commandQueuesOpenCL command queues.
[in]numEventsInWaitListNumber of events in the event wait list.
[in]eventWaitListEvent wait list.
[in]eventsEvent objects per each command queue that identify a particular kernel execution instance.
Returns
  • clblasSuccess on success;
  • clblasInvalidDevice if a target device does not support floating point arithmetic with double precision;
  • clblasInvalidValue if either offA or offB exceeds the size of the respective buffer object;
  • the same error codes as the clblasStrmm() function otherwise.