Freescale Semiconductor,inc.

Nxp Usa, Inc.

Robert N Ehrlich

Robert A McGowan

Brett W Murdock

Craig D Shaw

Michael Brian Schinzler

George E Baker

Alan Joseph Carlin

Matthew D Akers

Darrell L Carder

Kevin C Heuer

Petru Lauric

Amar Nath Deogharia

Rakesh Bakhshi

Michael L Olivarez

Stephen J Benzel

A method and information processing system provide trace compression for trace messages. In response to a branch of a conditional branch instruction having not been taken or having been taken, a flag of a history buffer is set or cleared. A trace address message is generated in response to a conditional indirect branch instruction being taken, wherein the trace address message includes address information indicating the destination address of the taken branch, and an index value indicating a corresponding flag of the history buffer. In response to a return from interrupt or return from exception instruction, a predicted return address is compared to an actual return address. A trace address message is generated in response to the predicted and actual return addresses not matching. A trace address message is not generated in response to the predicted and actual return addresses matching.

Debug method and device for handling exceptions and interrupts

Debug method and device for providing indexed trace messages

A program trace data compression mechanism in which execution of a variable length execution set (VLES) including multiple non-branch conditional instructions are traced in real-time in a manner that allows the instruction execution to be reconstructed completely by correlating the trace data with the traced binary code.

Predicate trace compression

The present disclosure provides system and method embodiments for generation of capture clock signals. A first and second test circuit receive a first test pattern and a functional clock signal. A first test clock control (TCC) circuit of the first test circuit generates a first capture clock signal that comprises a set of functional clock signal pulses generated according to a first clock pattern of the first test pattern. A second TCC circuit of the second test circuit generates a second capture clock signal that comprises the set of functional clock signal pulses generated according to the first clock pattern. The set of functional clock signal pulses of the second capture clock signal are staggered in time from the set of functional clock signal pulses of the first capture clock signal.

Scan testing with staggered clocks

A method and apparatus for an asynchronous multicore common debugging system is described. Debug signals from a plurality of processor cores are synchronized to a common timing domain. Processing completed within the plurality of processor cores during a common timing interval is tracked. A single debugging tool chain is utilized to provide debugging results in response to the tracking the processing completed within the plurality of processor cores during the common timing interval.

System and method for multicore processing

A heterogeneous multi-core integrated circuit includes first and second sets of processor cores and corresponding first and second test access ports (TAPs). The first and second TAPs are connected to corresponding first and second debug ports by way of corresponding first and second TAP controllers. A debug control circuit is connected between the first and second TAP controllers and the first and second debug ports. Based on external configuration signals, the debug control circuit configures the connections between the first and second TAP controllers and the first and second debug ports according to predetermined configuration modes, which allows flexibility in debugging the heterogeneous multi-core integrated circuit.

Heterogeneous multi-core integrated circuit and method for debugging same

A device is configured with components to enable debugging of the device's entry into and exit from a low power mode. The device includes: core logic, debug components, and a power management module (PMM). When the device exits a low power mode in which the states of the debug components are lost, the PMM prevents the core logic from resuming processing operations until the debug components have been re-configured to their prior states. The PMM either holds the core logic in reset or alternatively withholds power to the core logic. Reconfiguration of the debug components is initiated by a connected debugger, which can set one or more control and status (CS) register values within the device. The CS register values determine when the PMM prevents the core logic processing from resuming and when the PMM enables core logic processing to resume following the device's return from low power mode.

Debugger recovery on exit from low power mode

Each level of cache within a memory hierarchy of a device is configured with a cache results register (CRR). The caches are coupled to a debugger interface via a peripheral bus. The device is placed in debug mode, and a debugger forwards a transaction address (TA) of a dummy transaction to the device. On receipt of the TA, the device processor forwards the TA via the system bus to the memory hierarchy to initiate an address lookup operation within each level of cache. For each cache in which the TA hits, the cache controller (debug) logic updates the cache's CRR with Hit, Way, and Index values, identifying the physical storage location within the particular cache at which the corresponding instruction/data is stored. The debugger retrieves information about the hit/miss status, the physical storage location and/or a copy of the data via direct requests over the peripheral bus.

Cache result register for quick cache information lookup

A flexible interrupt controller circuit and methodology are provided which use an interrupt circuit () that multiplexes () a plurality of interrupt priority registers () based on the current context of the system that is identified in mode control selector (). By using the mode control selector () to selectively couple different priority level assignments to a priority encoding module (), context sensitive switching of the priority levels assigned to each interrupt request can be implemented with reduced latency. The context switch could be based on an OS context ID, power management modes, security modes, and other system defined modes where priority levels would differ. The selected priority level information is used to provide an interrupt request signal () which will cause an interrupt to occur in the data processing system.

Hardware managed context sensitive interrupt priority level control

A data processing system may comprise an initiator device having an output whose timing is referenced by a clock input alone corresponding to a first delay along a signaling path. The exemplary data processing system further may further comprise a target device having an input whose timing is referenced by a clock input alone corresponding to a second delay along the signaling path and a system bus interconnected between the initiator device and the target device within the signaling path. The exemplary data processing system may further comprise a dynamic timing bridge coupled to the system bus within the signaling path, wherein responsive to a control signal representative of at least one system characteristic, the dynamic timing bridge performs one selected from the group consisting of (i) inserting a cyclic latency within the signaling path and (ii) not inserting the cyclic latency within the signaling path.

Growing community of inventors

Round Rock, TX, United States of America

Robert N Ehrlich