Hello all:
I am working on collecting video stream from a FPGA board which generates image test pattern. The image pattern generated by the FPGA board is 800x600, line rate=48.1kHz, frame rate=72Hz, PCLK=50MHz. The timing is consistent with the output of OV5642. I am trying to modify the driver of mt9p031 to customize my current need. So far I have done something by commenting out all the I2C operation with the sensor. It works on my application before the DQBUF, but I get trouble in collecting image data from the FPGA after DQBUF is called.
Since no physical sensor is configured, what are the key factors that I should take care in the driver in order to match the image pattern from the FPGA? I realized that width, height, x_current and y_current are crucial to the kernel to successfully collect image data after the STREAMON is operated. Are there any other things that I should modify to match the coming image data? Do I need to configure VSYNC, HREF and PCLK in the kernel? I didn't find where I should configure them.
In addition, please let me know how to let kernel debug/err info appeared at the console. I find the debug info like v4l2_dbg(1, debug, &vpfe_dev->v4l2_dev, "vpfe_enum_input\n"), but I never see it at the console. Also the same in the API like Dmai_err1("%s does not support streaming i/o\n", attrs->captureDevice), I have to add a printf to debug.
Many thanks for all your generous help.
Jerry