libflame revision_anchor
Functions
FLA_Gemm_ct_blk_var2.c File Reference

(r)

Functions

FLA_Error FLA_Gemm_ct_blk_var2 (FLA_Obj alpha, FLA_Obj A, FLA_Obj B, FLA_Obj beta, FLA_Obj C, fla_gemm_t *cntl)
 

Function Documentation

◆ FLA_Gemm_ct_blk_var2()

FLA_Error FLA_Gemm_ct_blk_var2 ( FLA_Obj  alpha,
FLA_Obj  A,
FLA_Obj  B,
FLA_Obj  beta,
FLA_Obj  C,
fla_gemm_t cntl 
)
16{
17 FLA_Obj AT, A0,
18 AB, A1,
19 A2;
20
21 FLA_Obj CT, C0,
22 CB, C1,
23 C2;
24
25 dim_t b;
26
27 FLA_Part_2x1( A, &AT,
28 &AB, 0, FLA_BOTTOM );
29
30 FLA_Part_2x1( C, &CT,
31 &CB, 0, FLA_BOTTOM );
32
33 while ( FLA_Obj_length( AB ) < FLA_Obj_length( A ) ){
34
36
38 &A1,
39 /* ** */ /* ** */
40 AB, &A2, b, FLA_TOP );
41
43 &C1,
44 /* ** */ /* ** */
45 CB, &C2, b, FLA_TOP );
46
47 /*------------------------------------------------------------*/
48
49 /* C1 = alpha * A1 * B' + C1; */
51 alpha, A1, B, beta, C1,
52 FLA_Cntl_sub_gemm( cntl ) );
53
54 /*------------------------------------------------------------*/
55
57 /* ** */ /* ** */
58 A1,
59 &AB, A2, FLA_BOTTOM );
60
62 /* ** */ /* ** */
63 C1,
64 &CB, C2, FLA_BOTTOM );
65
66 }
67
68 return FLA_SUCCESS;
69}
FLA_Error FLA_Gemm_internal(FLA_Trans transa, FLA_Trans transb, FLA_Obj alpha, FLA_Obj A, FLA_Obj B, FLA_Obj beta, FLA_Obj C, fla_gemm_t *cntl)
Definition FLA_Gemm_internal.c:16
FLA_Error FLA_Cont_with_3x1_to_2x1(FLA_Obj *AT, FLA_Obj A0, FLA_Obj A1, FLA_Obj *AB, FLA_Obj A2, FLA_Side side)
Definition FLA_View.c:428
FLA_Error FLA_Repart_2x1_to_3x1(FLA_Obj AT, FLA_Obj *A0, FLA_Obj *A1, FLA_Obj AB, FLA_Obj *A2, dim_t mb, FLA_Side side)
Definition FLA_View.c:226
FLA_Error FLA_Part_2x1(FLA_Obj A, FLA_Obj *A1, FLA_Obj *A2, dim_t mb, FLA_Side side)
Definition FLA_View.c:76
dim_t FLA_Obj_length(FLA_Obj obj)
Definition FLA_Query.c:116
dim_t FLA_Determine_blocksize(FLA_Obj A_unproc, FLA_Quadrant to_dir, fla_blocksize_t *cntl_blocksizes)
Definition FLA_Blocksize.c:234
unsigned long dim_t
Definition FLA_type_defs.h:71
int i
Definition bl1_axmyv2.c:145
Definition FLA_type_defs.h:159

References FLA_Cont_with_3x1_to_2x1(), FLA_Determine_blocksize(), FLA_Gemm_internal(), FLA_Obj_length(), FLA_Part_2x1(), FLA_Repart_2x1_to_3x1(), and i.

Referenced by FLA_Gemm_ct().