SPRUJ28E November 2021 – September 2024 AM68 , AM68A , TDA4AL-Q1 , TDA4VE-Q1 , TDA4VL-Q1
The read requests sent to the system interconnect are pipelined and arbitrated in a round-robin scheme. The default arbitration scheme can be modified by setting the priority attribute of each pipeline as defined in the DSS0_VID_ATTRIBUTES[23] ARBITRATION register bit.
By default, all pipelines have the same priority (normal priority), which means all pipeline requests are treated in a round-robin order manner. If one or more pipelines require a higher number of requests going to the system interconnect, its priority can be moved up to high priority. In this case, the high-priority pipeline is granted access before any pipeline in normal priority. If more than one active pipeline is in high priority, then the behavior is the same as all active pipelines in normal priority. Normal active pipelines are not treated until all high active pipelines are finished. The ARBITRATION bit cannot be modified during the entire frame.
In addition to the priority bit-field, the MFLAG mechanism can also result in a higher priority for a pipeline. An MFLAG priority level is set, for all pipelines for which MFLAG bit is set, which is defined as the highest priority level for arbitration. For more details on MFLAG mechanism, see Section 12.6.3.6.7, DISPC DMA MFLAG Mechanism. Figure 12-316 shows the transition between the different priority levels.