In order to prevent the data overflow or underflow, the external processor should know when the buffer is full (writing case) or empty(reading case). Flags are used for this purpose.
There is a finite delay present between the point at which the buffers inside the FX3 become full/empty and actual assertion of flags (for example - Thread_0_DMA_READY ). The delay is different for different flags as mentioned in table 4 of AN65974.
Hence, there is a possibility that by the time the external processor sees the flag,it might have send an extra bytes to GPIF II (assume external processor is writing to FX3 GPIF II) due to the delay in assertion of flag.
This is where the partial flags coming into the picture. This flag should be configured in such a way that the external processor should know how many cycles it can read/write the data from/to the GPIF.
i.e the number of bytes that the external processor can read/write from/to the GPIF should be known to the external processor and stop the data transfer after the count. Later wait for de-assertion of the watermark or ready flag then do the data transfer.
To use water mark flags, you must configure them in GPIF Designer and set the water mark value using the API CyU3PGpifSocketConfigure. Note that we cannot set the value in the GPIF designer, hence we must use this API. Hence, we must configure the water mark flag in GPIF II designer and use this API to set water mark value.
Go through the section 8.2, 8.3 and 8.4 to understand this.
You may go through this thread - Re: FX3 Flag polarity and using partial flags
