AV renderer peripheral with dual inerrupt lines for staggered interrupts
View Patent ↗An audio-video (AV) playback chain and rendering peripheral device 300 for generating two or more interrupts 306, 308 with a programmable delay 304 between them. Such are configured to prevent two processors 310, 314 from racing to access the same system resources in their respective interrupt service routines.
1. A system for use in an audio-video (AV) playback chain, comprising:
an AV renderer peripheral for processing an input datastream;
a first interrupt service request output at a first interrupt line to be generated by a predetermined event occurring in the AV renderer peripheral, and providing for a first processor to begin a first interrupt service routine to service a first task for the AV renderer peripheral;
a second interrupt service request output providing for a second processor to begin a second interrupt service routine to service a second task for the AV renderer peripheral; and
a programmable delay timer configured to be triggered by the first interrupt service request output and to generate the second interrupt service request output at a second interrupt line after a delay relative to the first interrupt service request output, wherein the delay depends on a programmed register value.
2. The system of claim 1 , wherein: the programmable delay timer uses the programmed register value to count down peripheral clock ticks.
3. The system of claim 1 , wherein: the programmable delay timer uses the programmed register value to count down bytes in an AV-data buffer.
4. A method of AV-rendering in a system-on-chip (SoC) with at least two processor cores, comprising:
causing a first interrupt service request at a first interrupt line to be generated by a predetermined event occurring in an AV renderer peripheral, the first interrupt service request configured to trigger a first processor to begin a first interrupt service routine to service a first task for said AV renderer; and
timing a second interrupt service request at a second interrupt line to a second processor to begin a second interrupt service routine to service a second task for said AV renderer peripheral with a delay relative to the first interrupt service request; wherein the delay depends on a programmed register value; wherein race conditions are controlled and limited by the delay.
5. The method of claim 4 , further comprising: dividing the tasks of tuning, decoding, post-processing, and rendering between said first and second processors.
6. The method of claim 4 , wherein: the timing provides for an interleaving of accesses from said first and second processors through a MMIO register structure to memory.
7. A system-on-chip (SoC), comprising:
a first core processor providing for execution of a first interrupt service routine;
a second core processor providing for execution of a second interrupt service routine; and
an AV renderer peripheral providing for management of the first and second processors with:
a first interrupt service request at a first interrupt line generated by a predetermined event occurring in the AV renderer peripheral, the first interrupt service request configured to trigger the first processor to begin executing said first interrupt service routine wherein said first interrupt service routine is configured to service a first task for said AV renderer peripheral, and
a second interrupt service request, at a second interrupt line, to the second processor to begin said second interrupt service routine wherein said second interrupt service routine is configured to service a second task for the AV renderer peripheral, wherein the second interrupt service request is generated after a delay relative to the first interrupt service request, wherein the delay depends on a programmed register value.