For the same task, if the device is changed from FPGA to GPU, the task execution time increases from 3 to 73.The code is in the attachment. Please help me analyze the reason.
Thank you for posting in Intel Communities. Unfortunately, we couldn't find any code attached. Could you please re-attach it and share all the steps you've done?
We reproduced your issue from our end and we could observe that you've used -DFPGA_EMULATOR flag while compiling the code. By doing this, FPGA emulator executes the dpc++ device code on the CPU. The FPGA emulator device supports FPGA extensions such as FPGA pipes and fpga_reg but doesn't compile your code on FPGA hardware. The FPGA emulator's timing behavior is not correlated to that of the physical FPGA hardware. For example, an optimization that yields a 100x performance improvement on the FPGA may show no impact on the emulator performance. The emulator might show an unrelated increase or decrease. For more information on types of DPC++ FPGA compilation, please refer: https://www.intel.com/content/www/us/en/develop/documentation/oneapi-programming-guide/top/programming-interface/fpga-flow/types-of-dpc-fpga-compilation.html
If the above information resolves your issue, make sure to accept this as a solution. This would help others with similar issue. Thank you!
We assume that your issue is resolved. If you need any additional information, please post a new question as this thread will no longer be monitored by Intel.
