CP_DBCSR

Section can be repeated.

Benchmark and test the cp_dbcsr routines [Edit on GitHub]

Keywords

• ALPHA

• ALWAYS_CHECKSUM

• ASPARSITY

• ATYPE

• BETA

• BSPARSITY

• BS_K

• BS_M

• BS_N

• BTYPE

• CSPARSITY

• CTYPE

• DATA_TYPE

• FILTER_EPS

• K

• KEEPSPARSE

• M

• N

• NPROC

• N_LOOP

• TEST_TYPE

• TRANSA

• TRANSB

Keyword descriptions

ALPHA

Type: real
Default: 1.00000000E+000
Usage: ALPHA 2.0

Description: Multiplication factor [Edit on GitHub]

ALWAYS_CHECKSUM

Type: logical
Default: F
Lone keyword: T
Usage: ALWAYS_CHECKSUM

Description: perform a checksum after each multiplication [Edit on GitHub]

ASPARSITY

Type: real
Default: 0.00000000E+000
Usage: ASPARSITY 70

Description: Sparsity of A matrix [Edit on GitHub]

ATYPE

Type: string
Default: N
Usage: ATYPE N

Description: Matrix A type [Edit on GitHub]

BETA

Type: real
Default: 0.00000000E+000
Usage: BETA 1.0

Description: Product premultiplication factor [Edit on GitHub]

BSPARSITY

Type: real
Default: 0.00000000E+000
Usage: BSPARSITY 80

Description: Sparsity of B matrix [Edit on GitHub]

BS_K

Type: integer[ ]
Default: 1 13 2 15
Usage: BS_K 1 13 2 5

Description: Block sizes of inner dimension [Edit on GitHub]

BS_M

Type: integer[ ]
Default: 1 13 2 15
Usage: BS_M 1 13 2 5

Description: Row block sizes of C [Edit on GitHub]

BS_N

Type: integer[ ]
Default: 1 13 2 15
Usage: BS_N 1 13 2 5

Description: Column block sizes of C [Edit on GitHub]

BTYPE

Type: string
Default: N
Usage: BTYPE N

Description: Matrix B type [Edit on GitHub]

CSPARSITY

Type: real
Default: 0.00000000E+000
Usage: CSPARSITY 90

Description: Sparsity of C matrix [Edit on GitHub]

CTYPE

Type: string
Default: N
Usage: CTYPE N

Description: Matrix C type [Edit on GitHub]

DATA_TYPE

Type: enum
Default: REAL_8
Usage: DATA_TYPE real_8

Description: Data type of the matrices [Edit on GitHub]

Valid values:

• REAL_8 Real (Double Precision)

• COMPLEX_8 Complex (Double Precision)

FILTER_EPS

Type: real
Default: -1.00000000E+000
Usage: FILTER_EPS 1.0

Description: Threshold for on-the-fly and final filtering. [Edit on GitHub]

K

Type: integer
Default: 256
Usage: K 1024

Description: Inner dimension M [Edit on GitHub]

KEEPSPARSE

Type: logical
Default: F
Lone keyword: T
Usage: KEEPSPARSE

Description: Keep product sparse [Edit on GitHub]

M

Type: integer
Default: 256
Usage: M 1024

Description: Dimension 1 of C [Edit on GitHub]

N

Type: integer
Default: 256
Usage: N 1024

Description: Dimension 2 of C [Edit on GitHub]

NPROC

Type: integer[ ]
Default: 0
Usage: NPROC 128 16 1

Description: Number of processors to test [Edit on GitHub]

N_LOOP

Type: integer
Default: 10
Usage: N_LOOP 10

Description: Number of operations being timed (useful for small matrices). [Edit on GitHub]

TEST_TYPE

Type: enum
Default: MM
Usage: TEST_TYPE MM

Description: Which part of DBCSR is tested [Edit on GitHub]

Valid values:

• MM Run matrix multiplications

• BINARY_IO Run binary IO checks

TRANSA

Type: logical
Default: F
Lone keyword: T
Usage: TRANSA

Description: Transpose matrix A [Edit on GitHub]

TRANSB

Type: logical
Default: F
Lone keyword: T
Usage: TRANSB

Description: Transpose matrix B [Edit on GitHub]