trmm (USM Version)¶
Computes a matrix-matrix product where one input matrix is triangular and one input matrix is general.
Description¶
The trmm
routines compute a scalar-matrix-matrix product where one of
the matrices in the multiplication is triangular. The argument
left_right
determines if the triangular matrix, A
, is on the
left of the multiplication (left_right
= side::left
) or on
the right (left_right
= side::right
). Depending on
left_right
. The operation is defined as
or
where:
op(
A
) is one of op(A
) = A, or op(A
) =A
T, or op(A
) =A
H,alpha
is a scalar,A
is a triangular matrix, andB
is a general matrix.
Here B
is m
x n
and A
is either m
x m
or n
x n
, depending on left_right
.
API¶
Syntax¶
namespace oneapi::mkl::blas::column_major {
sycl::event trmm(sycl::queue &queue,
onemkl::uplo upper_lower,
onemkl::transpose transa,
onemkl::diag unit_diag,
std::int64_t m,
std::int64_t n,
T alpha,
const T* a,
std::int64_t lda,
T* b,
std::int64_t ldb,
const std::vector<sycl::event> &dependencies = {})
}
namespace oneapi::mkl::blas::row_major {
sycl::event trmm(sycl::queue &queue,
onemkl::uplo upper_lower,
onemkl::transpose transa,
onemkl::diag unit_diag,
std::int64_t m,
std::int64_t n,
T alpha,
const T* a,
std::int64_t lda,
T* b,
std::int64_t ldb,
const std::vector<sycl::event> &dependencies = {})
}
trmm
supports the following precisions and devices:
T |
Devices Supported |
---|---|
|
Host, CPU, and GPU |
|
Host, CPU, and GPU |
|
Host, CPU, and GPU |
|
Host, CPU, and GPU |
Input Parameters¶
- exec_queue
The queue where the routine should be executed.
- left_right
Specifies whether
A
is on the left side of the multiplication (side::left
) or on the right side (side::right
). See Data Types for more details.- upper_lower
Specifies whether the matrix
A
is upper or lower triangular. See Data Types for more details.- transa
Specifies op(
A
), the transposition operation applied toA
. See Data Types for more details.- unit_diag
Specifies whether
A
is assumed to be unit triangular (all diagonal elements are 1). See Data Types for more details.- m
Specifies the number of rows of
B
. The value ofm
must be at least zero.- n
Specifies the number of columns of
B
. The value ofn
must be at least zero.- alpha
Scaling factor for the matrix-matrix product.
- a
Pointer to input matrix
A
. Must have size at leastlda
*m
ifleft_right
=side::left
, orlda
*n
ifleft_right
=side::right
. See Matrix and Vector Storage for more details.- lda
Leading dimension of
A
. Must be at leastm
ifleft_right
=side::left
, and at leastn
ifleft_right
=side::right
. Must be positive.- b
Pointer to input/output matrix
B
. It must have size at leastldb
*n
if column major layout is used to store matrices or at leastldb
*m
if row major layout is used to store matrices. See Matrix and Vector Storage for more details.- ldb
Leading dimension of
B
. It must be positive and at leastm
if column major layout is used to store matrices or at leastn
if column major layout is used to store matrices.- dependencies
List of events to wait for before starting computation, if any. If omitted, defaults to no dependencies.
Output Parameters¶
- b
Pointer to the output matrix, overwritten by
alpha
*op(A
)*B
oralpha
*B
*op(A
).Note
If
alpha
= 0, matrixB
is set to zero, andA
andB
do not need to be initialized at entry.
Return Values¶
Output event to wait on to ensure computation is complete.