SetDoubleBuffer
Function Usage
Determines whether to enable double buffer for A, B, C, and bias, and whether to enable ND2NZ or NZ2ND conversion. This API is used for internal optimization of the tiling function.
This API is reserved and not supported in the current version.
Prototype
1 | int32_t SetDoubleBuffer(bool a, bool b, bool c, bool bias, bool transND2NZ = true, bool transNZ2ND = true) |
Parameters
Parameter |
Input/Output |
Description |
|---|---|---|
a |
Input |
Determines whether to enable double buffer for matrix A. |
b |
Input |
Determines whether to enable double buffer for matrix B. |
c |
Input |
Determines whether to enable double buffer for matrix C. |
bias |
Input |
Determines whether to enable double buffer for the bias matrix. |
transND2NZ |
Input |
Sets whether ND2NZ is required. |
transNZ2ND |
Input |
Sets whether NZ2ND is required. |
Returns
The value -1 indicates that the setting fails, and the value 0 indicates that the setting is successful.
Precautions
None
Parent topic: Matmul Tiling Class