caculate

Performs calculation using basic operators, including cryptographic matrix addition, subtraction, and multiplication operators.

int calculate(
    DG_TeeCtx *dgTeeCtx,
    DG_AlgorithmsType type,
    DG_MpcMatShareSet *shareSet,
    DG_MpcMatShare **share
)

Parameters

Parameter	Description	Value Range	Input/Output
dgTeeCtx	KCAL context	Context initialized in section initTeeCtx.	Input
type	Operator enumeration: DG_AlgorithmsType	MATRIX_ADD: matrix addition MATRIX_SUB: matrix subtraction MATRIX_MUL: matrix multiplication	Input
shareSet	The type of the structure after concatenating the shares: DG_MpcMatShareSet	The value is not null.	Input
share	The type of the calculation result: DG_MpcMatShare	It is null as a single pointer and cannot be null as a double pointer.	Output

**Table 1** DG_MpcMatShareSet structure
Member	Type	Description
shareSet	DG_MpcMatShare	Shares of each participant
size	unsigned long	Number of participants

**Table 2** DG_MpcMatShare structure
Member	Type	Description
dataShare	Share*	Share structure
row	unsigned long	Number of rows in a matrix
col	unsigned long	Number of columns in a matrix
shareType	DG_ShareType	Share type

Return Values

Success: 0 is returned.
Failure: The error code is returned.

Error Codes

Error Code	Value	Description	Remarks
DG_ERR_MPC_TEE_INVALID_PARAM	4501	Invalid parameter	-
DG_ERR_MPC_TEE_INVALID_ALGORITHM_TYPE	4502	Invalid algorithm type	-
DG_ERR_MALLOC_FAIL	51	Memory allocation failure	-
DG_ERR_MPC_TEE_DIV_IS_ZERO	4512	Division by zero error	-
DG_FAILURE	4501	Operator execution failure	Failed to verify the parameters. Failed to split data into shares. Failed to receive data from the other node.

Dependency

dgTeeCtx depends on the successful initialization of initTeeCtx.
The input parameter depends on the result of a successful makeShare call.
#include "data_guard_mpc.h": required header file

Example

#include  "data_guard_mpc.h" 
// Compute party
DG_AlgorithmsType type = MATRIX_MUL;
std::unique_ptr<DG_MpcMatShareSet> shares = std::make_unique<DG_MpcMatShareSet>();
std::unique_ptr<DG_MpcMatShare[]> share_datas = std::make_unique<DG_MpcMatShare[]>(2);
share_datas[0] = *share2;
share_datas[1] = *share1;
shares->shareSet = share_datas.get();
shares->row = 2;
shares->col = 2;
DG_MpcMatShare *share_out = nullptr;
res = aritOpts.calculate(dgTee, type, shares.get(), &share_out);

Note: The calculation uses the matrix multiplication operator as an example, and shares 1 and 2 are for reference and cannot be executed.

Running result: The shares are not null and res is 0.

Call the calculate API using the dgTeeCtx output from the successful initTeeCtx call.
The share output indicates the data share of a single party. You need to call the makeShare API multiple times and concatenate the DG_MpcMatShare data from both parties into DG_MpcMatShareSet. The concatenation order is crucial. For example, after compute node 0 acquires share a, it calls the matrix multiplication operator to calculate matrix a multiplying matrix b (a originates from compute node 1 and b from compute node 0), and then places share a at bit 0 and share b at bit 1 in DG_MpcMatShareSet.
Only two-party operations are supported, with one party set to node 0 and the other to node 1.

Parent topic: Matrix Arithmetic Operators