Hi
Thank you for your answer.I had saw these discussion andi am sorry about that my question is unclearly.My question is that which is the better method?use the V2L framework to capture and process the sensor data or write a no-standard linux driver , I am care about it very much , because of after capture the data, I need do a lot of computing, max to 120fps, and I am pressed for time, I must choose a fast way to do it, But on my experience unable to decide. Can you help me?