International Business Machines Corporation

Manfred Walz

Benedikt Geukes

Matteo Michel

Thomas Buechner

Heiko Michel

Matthias Klein

Andreas Wagner

Rolf Fritz

Holger Horbach

Gerhard Zilles

Martin Doerr

Martin Eckert

Markus Helms

Thomas Schlipf

Kirk David Lamb

Andreas Koenig

Christopher S Smith

Ulrich Olderdissen

Carsten Schmitt

Automatic detection of clock grid misalignments and automatic realignment including loading a test pattern into a first storage element on a first clock grid on a circuit; scanning the test pattern from the first storage element on the first clock grid to a second storage element on a second clock grid on the circuit; reading the scanned test pattern from the second storage element on the second clock grid; evaluating the scanned test pattern for errors; and in response to detecting an error in the scanned test pattern, triggering an alignment of the first clock grid and the second clock grid.

Automatic detection of clock grid misalignments and automatic realignment

A method, circuit, and design structure for an on-chip sequence profiler involves a programmable matrix and FSM in a logic circuit, a first latch receiving a scan path bit sequence and a first clock signal and generating a first output to control a select input of a first multiplexer, a first multiplexer selecting among functional path bit sequences and outputting a bit sequence to a second latch, a second latch receiving the bit sequence from the first multiplexer and a second clock signal via a second multiplexer and outputting a bit sequence to a logic circuit, a logic circuit receiving the bit sequence from the second latch and outputting a clock control signal to the clock and a second selector control signal to the second multiplexer, and a second multiplexer receiving a second clock signal from the clock and outputting the second clock signal to the second latch.

On-chip sequence profiler

The present invention relates to a method for a self-boot of an electronic device, wherein an external component is accessible through an interface of the electronic device (), the method comprising, determining a boot mode for booting the electronic device, wherein the determined boot mode is defined as a secure boot mode; disabling the interface, thereby prohibiting access to the component through the interface, thereby defining a secure state of the electronic device; loading a first code comprising a sequence of executable instructions to be executed for booting the electronic device; loading a second code, the second code being encrypted; and decrypting the second code and executing the second code, thereby enabling the interface, and switching the electronic device from the secure state to a debugging state.

Method for secure self-booting of an electronic device

A scan circuit (JTAG 1149 extension) for a microprocessor utilizes transport logic and scan chains which operate at a faster clock speed than the external JTAG clock. The transport logic converts the input serial data stream (TDI) into input data packets which are sent to scan chains, and converts output data packets into an output data stream (TDO). The transport logic includes a deserializer having a sliced input buffer, and a serializer having a sliced output buffer. The scan circuit can be used for testing with boundary scan latches, or to control internal functions of the microprocessor. Local clock buffers can be used to distribute the clock signals, controlled by thold signals generated from oversampling of the external clock. The result is a JTAG scanning system which is not limited by the external JTAG clock speed, allowing multiple internal scan operations to complete within a single external JTAG cycle.

Method for JTAG-driven remote scanning

A technique for controlling scan access of multiple scan devices (including or more slave scan devices and a master scan device) to a scan chain includes sending, by a requesting slave scan device included in the one or more slave scan devices, a first request for access to the scan chain to the master scan device. The master scan device and the one or more slave scan devices are connected to the scan chain. The technique also includes receiving, at the requesting slave scan device, an evaluation result from the master scan device and accessing, by the requesting slave scan device, the scan chain in response to the evaluation result indicating access granted. Finally, the technique includes sending, by the requesting slave scan device, one or more second requests for access to the scan chain to the master scan device in response to the evaluation result indicating access denied.

Efficiency of compression of data pages

Controlling scan access to a scan chain

Apparatus for JTAG-driven remote scanning

In some embodiments, a computer-implemented method includes receiving, in a processor, a device description code identifying components of a device and connections between the components, wherein some of the components and connections form boundary cells used for testing the device. The method can include processing, in the processor, the device description code to determine that the components and the connections meet a standard governing components and connections necessary for the boundary cells. The method can also include traversing the connections between the components to determine that the connections meet the standard, and reporting, via one or more output devices, that the device complies with the standard.

Verifying and detecting boundary scan cells to input/output mapping

Additional circuitry is included in an input cell design structure for an integrated circuit to detect and report transitions on an input that was expected to be stable, and to store that event for later analysis. Two or more modified input cells may have their error indications daisy-chained together to minimize additional routing. The storage elements may be included in a scan chain to allow for isolation of which input had the unexpected transition.

Detecting an unstable input to an IC

Methods and corresponding test systems for generating a chip facility waveform from a series of chip snapshots. The methods including, (i) testing an integrated chip multiple times, each time increasing a clockstop delay delaying a clockstop generated by triggered error condition each time determining the state of state holding elements of the integrated circuit and (ii) testing an integrated circuit chip one time to generate a error condition and determining multiple times the states of state holding elements of the integrated circuit based on previous states of the state holding elements.

Method and system for generating an integrated circuit chip facility waveform from a series of chip snapshots

A method and circuits for monitoring and detecting an error in the static pervasive signals applied to input/output pins of an integrated circuit during functional operation of the integrated circuit. The method and circuits provide a signal signature of each of one or more groups of the static pervasive signals and then monitoring the signal signature for any change of logic level.

Method and device to detect failure of static control signals

A non-obtrusive activity monitor is proposed for advantageously monitoring and tracing disjunct, concurrent computer system operations in heavily queued computer systems. For each traced and pending computer system operation, the monitor uses a hardware implementation of an event triggered operation graph to trace the path of the computer system operation through the computer system. For each followed path, a unique signature is generated that significantly reduces the amount of trace data to be stored. In a preferred embodiment, the trace information is stored together with a time stamp for debugging and measuring queuing effects and timing behavior in a computer system.

Operation graph based event monitoring system

A data mechanism having a random access memory (RAM) which has a plurality of groups of memory chips, each group being divisible into two equally sized chip sets. Each group of memory chips is addressed by a first address and each individual memory chip is addressed by a second address. The random access memory contains stored data. A cache, connected to the RAM, stores a portion of data stored in the RAM and is accessed by a cache address for separately reading requested data therefrom. The cache provides a cache miss signal when it does not contain the requested data. A CPU, connected to the cache and the RAM, receives the cache miss signal and provides responsive thereto, a starting address to the random access memory for starting a block transfer from the random access memory to the cache in two shots. The starting address includes the first address and the second address. The starting address identifies the group and individual chip within the group which contains the first bit which, when attempted to be read from the cache, caused the cache miss signal. A decoder, connected to the CPU and the random access memory, receives the starting address from the CPU and enables a first block data transfer from a first chip set in a first shot of the two shots starting from said first bit which caused the cache miss signal, and further enables a second block data transfer from a second chip set in a second of the shots.

Growing community of inventors

Boeblingen, Germany

Manfred Walz