Cascade Halt of an Application Across Configurable Unit Arrays in a Reconfigurable Processor

A reconfigurable processor, comprising:

• an internal network;
• a plurality of arrays of configurable units coupled to the internal network, the plurality of arrays including a first array of configurable units and a second array of configurable units;
• one or more registers including a first application ID field associated with the first array of configurable units and a second application ID field associated with the second array of configurable units;
• a hang detection circuit configured to detect a hang condition in execution of a first application on the first array of configurable units;
• and
• a network recovery circuit coupled to the internal network and configured to:
• cause, in response to the hang condition, execution of the first application on the first array of configurable units to be halted;
• send an interrupt to a host computer over an external communication link to indicate that the first array of configurable units is hung;
• compare a first value of the first application ID field to a second value of the second application ID field;
• and
• cause execution on the second array of configurable units to be halted in response to the first value being equal to the second value, which signifies that the second array of configurable units is also assigned to the first application.

A non-transitory computer-readable medium storing instructions that, when executed by a host computer coupled to a reconfigurable processor through an external communication link, cause the host computer to perform operations comprising:

• sending, to a first array of configurable units in the reconfigurable processor over the external communication link, first configuration information for a first portion of a first application, and setting a first application ID field associated with the first array of configurable units in the reconfigurable processor to a value indicative of the first application;
• sending, to a second array of configurable units in the reconfigurable processor over the external communication link, second configuration information for a second portion of the first application, and setting a second application ID field associated with the second array of configurable units in the reconfigurable processor to the value indicative of the first application;
• initiating execution of the first application using at least the first array of configurable units and the second array of configurable units in the reconfigurable processor;
• receiving, over the external communication link, an interrupt from the reconfigurable processor indicating that the first array of configurable units is hung, the interrupt being sent in response to a network recovery circuit of the reconfigurable processor detecting that the first array of configurable units is hung and causing execution on at least the first array of configurable units to be halted;
• and
• determining, in response to receiving the interrupt, that the second array of configurable units in the reconfigurable processor is assigned to the first application and that execution on the second array of configurable units has been halted by the network recovery circuit based on the second application ID field having the value indicative of the first application.