Why does the Nios® V/g processor return inaccurate floating-point calculation results when the Floating-Point Unit is enabled?

Description

Due to a problem in the Quartus® Prime Standard Edition Software version 24.1, the Nios® V/g processor might return inaccurate floating-point calculation results when the Floating-Point Unit (FPU) is enabled.

This problem is found in Max® 10 FPGA devices only. Other Altera® FPGA devices are not affected.

This is because there is a problem in the FPU module, which generates spurious signals affecting the calculation results.

Resolution

A patch is available to fix this problem for the Quartus® Prime Standard Edition Software version 24.1. 

Download and install patch 0.03 from the following links:

• Quartus® Prime Standard Edition Software v24.1 Patch 0.03

This problem is scheduled to be fixed in a future release of the Quartus® Prime Standard Edition Software.