Embodiments herein describe binning and placement techniques for assembling a multi-die device to improve yield when a customer requests a high performance feature from the device. For example, the multi-die device may include multiple dies that are interconnected to form a single device or package. In one embodiment, the multiple dies are the same semiconductor die (e.g., have the same circuit layout) which are disposed on a common interposer or stacked on each other. The multi-die device can then be attached to a printed circuit board (PCB). Although the dies in the multi-die device may each include the same feature (e.g., a PCIe interface, SerDes interface, transmitter, memory interface, etc.), the multi-die device is assembled so that not all of the dies have a feature that satisfies the high performance requested by the customer. That is, at least one of the die includes a lower performance feature.

Techniques facilitating determination and correction of physical circuit event related errors of a hardware design are provided. A system can comprise a memory that stores computer executable components and a processor that executes computer executable components stored in the memory. The computer executable components can comprise a simulation component that injects a fault into a latch and a combination of logic of an emulated hardware design. The fault can be a biased fault injection that can mimic an error caused by a physical circuit event error vulnerability. The computer executable components can also comprise an observation component that determines one or more paths of the emulated hardware design that are vulnerable to physical circuit event related errors based on the biased fault injection.

A method for detecting errors of a first field-programmable gate array (FPGA) program includes: receiving, by a monitoring program executed on a processor connected to an FPGA on which the first FPGA program is executed, a signal value read out from the first FPGA program; and comparing, by the monitoring program executed on the processor, the signal value to a reference value from a source other than the first FPGA program in order to detect errors of the first FPGA program.

A computer-implemented method for automated generation of test layouts for verifying a DRC deck. The method comprises receiving a first layout (L1) comprising one or more polygon shapes (P1) defined by a plurality of polygon parameters (W1,H1). Design rules (R1,R2) are received comprising inequality constraints (C) on the polygon parameters (W1,H1). A second layout (L2) is calculated by applying a random change (W12) of value to at least one of the polygon parameters (W1) of the first layout (L1). A third layout (L3) is calculated by varying values of the polygon parameters (W1,H1) of the second layout (L2) until a respective slack (S1,S2) of the polygon parameters (W1,H1) with respect to one or more of the parameter boundaries (B1,B2) defined by the constraints is minimized. The third layout (L3) may be stored as candidate test layout.

An apparatus, including: a deterministic monitored device; an interconnect to communicatively couple the monitored device to a support circuit; a super queue to queue transactions between the monitored device and the support circuit, the super queue including an operational segment and a shadow segment; a debug data structure; and a system management agent to monitor transactions in the operational segment, log corresponding transaction identifiers in the shadow segment, and write debug data to the debug data structure, wherein the debug data are at least partly based on the corresponding transaction identifiers.

A computer system includes a hardware accelerator and host processor. The hardware accelerator executes a simulation of a first logical model according to a plurality of simulation cycles. The host processor determines a fault checkpoint based on a logic fault that occurs in response to executing the simulation. The host processor verifies removal of the logic fault based on rerunning the simulation from the fault checkpoint.

A data reading device and a data reading method for design-for-testing are provided. The data reading device includes a buffer and a data serialization circuit. The data serialization circuit receives a clock positive edge-triggered signal, a clock negative edge-triggered signal, a trigger mask signal, and test data. The data serialization circuit masks one of the clock positive edge-triggered signal and the clock negative edge-triggered signal according to the trigger mask signal, and provides a part of the test data to an output terminal of the data serialization circuit as an output signal of the data reading device according to the unmasked one of the clock positive edge-triggered signal and the clock negative edge-triggered signal. Thus, a data valid window of the test data can be increased.

A data reading device and a data reading method for design-for-testing are provided. The data reading device includes a buffer and a data serialization circuit. The data serialization circuit receives a clock positive edge-triggered signal, a clock negative edge-triggered signal, a trigger mask signal, and test data. The data serialization circuit masks one of the clock positive edge-triggered signal and the clock negative edge-triggered signal according to the trigger mask signal, and provides a part of the test data to an output terminal of the data serialization circuit as an output signal of the data reading device according to the unmasked one of the clock positive edge-triggered signal and the clock negative edge-triggered signal. Thus, a data valid window of the test data can be increased.

Various examples of a circuit and a technique for testing the circuit are disclosed herein. In an example, the circuit includes a data input coupled to a scan multiplexer and a path select multiplexer. The circuit further includes a scan-in input coupled to the scan multiplexer and to receive a value of a scan pattern. The circuit further includes a scan latch to store the value that has an input coupled to the scan multiplexer and an output coupled to the path select multiplexer. The scan multiplexer selects a first signal from the data input and the scan-in input and provides the first signal to the input of the scan latch. The path select multiplexer selects a second signal from the data input and the output of the scan latch and provides the second signal to a data output of the circuit.

An information processing apparatus includes a processor including hardware. The processor extracts neighboring nodes in two or more different extraction ranges for each node constituting input data of a graph structure. The processor calculates an anomaly score representing a degree of anomaly of the node for each extraction range based on a representation of a combination of the node and the neighboring nodes. The processor records each calculated anomaly score in a storage. The processor selects a maximum anomaly score among the anomaly scores recorded in the storage. The processor determines an anomaly node in the input data of the graph structure based on the selected maximum anomaly score. The processor outputs information of the anomaly node.