CP_FM_GEMM

Section can be repeated.

Benchmark and test the cp_fm_gemm routines by multiplying C=A*B [Edit on GitHub]

Keywords

Keyword descriptions

FORCE_BLOCKSIZE

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

Description: Forces the blocksize, even if this implies that a few processes might have no data

[Edit on GitHub]

GRID_2D

Type: integer[2]
Default: 1 1
Usage: GRID_2D 64 16

Description: Explicitly set the blacs 2D processor layout. If the product differs from the number of MPI ranks, it is ignored and a default nearly square layout is used.

[Edit on GitHub]

K

Type: integer
Default: 256
Usage: K 1024

Description: Dimension 1 of C

[Edit on GitHub]

M

Type: integer
Default: 256
Usage: M 1024

Description: Inner dimension M

[Edit on GitHub]

N

Type: integer
Default: 256
Usage: N 1024

Description: Dimension 2 of C

[Edit on GitHub]

NCOL_BLOCK

Type: integer
Default: 32
Usage: NCOL_BLOCK 64

Description: block_size for cols

[Edit on GitHub]

NROW_BLOCK

Type: integer
Default: 32
Usage: nrow_block 64

Description: block_size for rows

[Edit on GitHub]

N_LOOP

Type: integer
Default: 10
Usage: N_LOOP 10

Description: Number of cp_fm_gemm operations being timed (useful for small matrices).

[Edit on GitHub]

ROW_MAJOR

Type: logical
Default: T
Lone keyword: T
Usage: ROW_MAJOR .FALSE.

Description: Use a row major blacs grid

[Edit on GitHub]

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]