Forum Discussion
Altera_Forum
Honored Contributor
8 years agoThis is how I do it:
1. Install the FPGA in a PCI-E slot in the host machine and then turn the machine on. 2. Program the FPGA and the on-board flash with the sof file provided by the manufacturer using JTAG, and set the FPGA to boot from the flash. 3. Turn off the host machine completely and turn it back on. 4. Set the environmental variable path for OpenCL in bashrc. 5. Install the FPGA driver (aocl install). 6. Run diagnose command (aocl diagnose). For the two boards listed, you have to choose the correct one manually for compiling OpenCL kernels. Your command will be "aoc --board *board_variant*". For diagnose, this doesn't matter.