Developer Reference for Intel® oneAPI Math Kernel Library for Fortran
p?gemm
Computes a scalar-matrix-matrix product and adds the result to a scalar-matrix product for distributed matrices.
Syntax
call psgemm ( transa , transb , m , n , k , alpha , a , ia , ja , desca , b , ib , jb , descb , beta , c , ic , jc , descc )
call pdgemm ( transa , transb , m , n , k , alpha , a , ia , ja , desca , b , ib , jb , descb , beta , c , ic , jc , descc )
call pcgemm ( transa , transb , m , n , k , alpha , a , ia , ja , desca , b , ib , jb , descb , beta , c , ic , jc , descc )
call pzgemm ( transa , transb , m , n , k , alpha , a , ia , ja , desca , b , ib , jb , descb , beta , c , ic , jc , descc )
Include Files
mkl_pblas.h
Description
The p?gemm routines perform a matrix-matrix operation with general distributed matrices. The operation is defined as sub( C ) := alpha *op(sub( A ))*op(sub( B )) + beta *sub( C ),
where:
op(x) is one of op(x) = x , or op(x) = x' ,
alpha and beta are scalars,
sub(A)=A(ia:ia+m-1, ja:ja+k-1) , sub(B)=B(ib:ib+k-1, jb:jb+n-1) , and sub(C)=C(ic:ic+m-1, jc:jc+n-1) , are distributed matrices.
Input Parameters
transa
(global) CHARACTER*1 . Specifies the form of op(sub(A)) used in the matrix multiplication:
if transa = 'N' or 'n' , then op(sub(A)) = sub(A) ; if transa = 'T' or 't' , then op(sub(A)) = sub(A)' ; if transa = 'C' or 'c' , then op(sub(A)) = sub(A)' .
transb
(global) CHARACTER*1 . Specifies the form of op(sub(B)) used in the matrix multiplication:
if transb = 'N' or 'n' , then op(sub(B)) = sub(B) ; if transb = 'T' or 't' , then op(sub(B)) = sub(B)' ; if transb = 'C' or 'c' , then op(sub(B)) = sub(B)' .
m
(global) INTEGER . Specifies the number of rows of the distributed matrices op(sub(A)) and sub(C) , m ≥ 0.
n
(global) INTEGER . Specifies the number of columns of the distributed matrices op(sub(B)) and sub(C) , n ≥ 0.
The value of n must be at least zero.
k
(global) INTEGER . Specifies the number of columns of the distributed matrix op(sub(A)) and the number of rows of the distributed matrix op(sub(B)) .
The value of k must be greater than or equal to 0.
alpha
(global) REAL for psgemm
DOUBLE PRECISION for pdgemm COMPLEX for pcgemm DOUBLE COMPLEX for pzgemm Specifies the scalar alpha . When alpha is equal to zero, then the local entries of the arrays a and b corresponding to the entries of the submatrices sub(A) and sub(B) respectively need not be set on input.
a
(local) REAL for psgemm
DOUBLE PRECISION for pdgemm COMPLEX for pcgemm DOUBLE COMPLEX for pzgemm Array, size lld_a by kla , where kla is LOCc(ja+k-1) when transa = 'N' or ‘n’ , and is LOCq(ja+m-1) otherwise. Before entry this array must contain the local pieces of the distributed matrix sub(A) .
ia , ja
(global) INTEGER . The row and column indices in the distributed matrix A indicating the first row and the first column of the submatrix sub(A) , respectively
desca
(global and local) INTEGER array of dimension 9. The array descriptor of the distributed matrix A .
b
(local) REAL for psgemm
DOUBLE PRECISION for pdgemm COMPLEX for pcgemm DOUBLE COMPLEX for pzgemm Array, size lld_b by klb , where klb is LOCc(jb+n-1) when transb = 'N' or ‘n’ , and is LOCq(jb+k-1) otherwise. Before entry this array must contain the local pieces of the distributed matrix sub(B) .
ib , jb
(global) INTEGER . The row and column indices in the distributed matrix B indicating the first row and the first column of the submatrix sub(B) , respectively
descb
(global and local) INTEGER array of dimension 9. The array descriptor of the distributed matrix B .
beta
(global) REAL for psgemm
DOUBLE PRECISION for pdgemm COMPLEX for pcgemm DOUBLE COMPLEX for pzgemm Specifies the scalar beta . When beta is equal to zero, then sub(C) need not be set on input.
c
(local) REAL for psgemm
DOUBLE PRECISION for pdgemm COMPLEX for pcgemm DOUBLE COMPLEX for pzgemm Array, size ( lld_a , LOCq(jc+n-1) ). Before entry this array must contain the local pieces of the distributed matrix sub(C) .
ic , jc
(global) INTEGER . The row and column indices in the distributed matrix C indicating the first row and the first column of the submatrix sub(C) , respectively
descc
(global and local) INTEGER array of dimension 9. The array descriptor of the distributed matrix C .
Output Parameters
- c
-
Overwritten by the m -by- n distributed matrix alpha*op(sub(A))*op(sub(B)) + beta*sub(C) .