Operation of floating point data of imx8qxp

[zhengguang_yang]

Hi 

We wrote a shell script on imx8qxp to perform floating point operations, and found that the CPU utilization rate reached more than 36%.

I would like to ask if there is any method for floating point operation, and the CPU utilization rate will not be too high.

Will it improve if you switch to hard floating-point? If yes, how do I change it to a hard floating point?

We use Yocto version 3.0 and kernel version 5.4.70

[Rita_Wang]

Sorry you need to realized yourself, or you can try our newest BSP to see the test result.