The present disclosure provides a static random access memory (SRAM) cell. The SRAM cell includes first and second inverters cross-coupled for data storage, each inverter including at least one pull-up device and at least two pull-down devices; at least four pass gate devices configured with the two cross-coupled inverters; at least two ports coupled with the at least four pass-gate devices for reading and writing; a first contact feature contacting first two pull-down devices (PD-11 and PD-12) of the first inverter; and a second contact feature contacting second two pull-down devices (PD-21 and PD-22) of the second inerter.

The present disclosure includes an integrated circuit comprising a first pair of complementary transistors configured in series, a second pair of complementary transistors configured in series, and at least one charge extraction transistor having a gate coupled to a first potential, a source coupled to a second potential, and a drain coupled to a data storage node of one of the first or second pairs of complementary transistors. The first potential and second potential bias the at least one charge extraction transistor in a nonconductive state. The drain of the at least one charge extraction transistor is formed in a doped material shared with a drain of a transistor of the first or second pairs of complementary transistors.

An in-memory computation circuit includes a memory array with SRAM cells connected in rows by word lines and in columns by bit lines. Body bias nodes of the transistors in each SRAM cell are biased by a modulated body bias voltage. A row controller circuit simultaneously actuates word lines in parallel for an in-memory compute operation. A column processing circuit processes analog voltages developed on the bit lines in response to the simultaneous actuation to generate a decision output for the in-memory compute operation. A voltage generator circuit switches the modulated body bias voltage from a non-negative voltage level to a negative voltage level during the simultaneous actuation. The negative voltage level is adjusted dependent on integrated circuit process and/or temperature conditions in order to optimize protection against unwanted memory cell data flip.

An in-memory computation circuit includes a memory array with SRAM cells connected in rows by word lines and in columns by bit lines. Each row includes a word line drive circuit powered by an adaptive supply voltage. A row controller circuit simultaneously actuates word lines in parallel for an in-memory compute operation. A column processing circuit processes analog voltages developed on the bit lines in response to the simultaneous actuation to generate a decision output for the in-memory compute operation. A voltage generator circuit generates the adaptive supply voltage for powering the word line drive circuits during the simultaneous actuation. A level of the adaptive supply voltage is modulated dependent on integrated circuit process and/or temperature conditions in order to optimize word line underdrive performance and inhibit unwanted memory cell data flip.

The invention introduces a method for verifying memory interface, performed by a processing unit, to include: driving a physical layer of a memory interface to pull-high or pull-low a signal voltage on each Input-Output (IO) pin thereof to a preset level according to a setting; obtaining a verification result corresponding to each IO pin from the memory interface; and storing each verification result in a static random access memory (SRAM), thereby enabling a testing host to obtain each verification result of the SRAM through a test interface. The testing host may examine each verification result to know whether any unexpected error has occurred in signals on the IO pins of the memory interface.

A circuit for mitigating single-effect-transients (SETs) comprising: a first sub-circuit comprising a first p-type transistor arrangement configured to generate a first output and a first n-type transistor arrangement configured to generate a second output; and a second sub-circuit comprising a connecting p-type transistor arrangement and a connecting n-type transistor arrangement connected in series, wherein the first output and the second output are electrically coupled to each other through the second sub-circuit.

An electronic circuit comprising an SRAM memory, a control unit, an error detection and correction module and a scrubbing module. The electronic circuit further comprises an integrated SEU monitor of the SRAM memory. The SEU monitor does not use standalone or specialized SRAM memories or particle detectors. Rather, the same SRAM memory that is used for the main operation as a storage element of the electronic circuit serves simultaneously as detector for the SEU monitor. The proposed SEU monitor enables real-time monitoring of the SEU rate in order to detect early the high radiation levels and apply appropriate hardening measures. Furthermore, a method for monitoring an SEU rate and determining permanent faults in an electronic circuit is suggested.

A semiconductor chip may include a memory, a power supply line, a noise generator and a switch. The power supply line may include first and second power supply line portions. The power supply line may be configured to provide a power supply signal through each of the first power supply line portion and the second power supply line portion. The noise generator may be connected to the second power supply line portion. The noise generator may be configured to receive the power supply signal from the second power supply line portion, and output a noisy power supply signal based on the power supply signal. The switch may be coupled to the memory, the first power supply line portion, and the noise generator. The switch may be configured to selectively electrically connect the memory to one of the first power supply line portion and the noise generator.

Examples described herein relate to a decision tree computation system in which a hardware accelerator for a decision tree is implemented in the form of an analog Content Addressable Memory (a-CAM) array. The hardware accelerator accesses a decision tree. The decision tree comprises of multiple paths and each path of the multiple paths includes a set of nodes. Each node of the decision tree is associated with a feature variable of multiple feature variables of the decision tree. The hardware accelerator combines multiple nodes among the set of nodes with a same feature variable into a combined single node. Wildcard values are replaced for feature variables not being evaluated in each path. Each combined single node associated with each feature variable is mapped to a corresponding column in the a-CAM array and the multiple paths of the decision tree to rows of the a-CAM array.

A memory array for storing odd and even data bits of data words in alternate sub-banks to reduce multi-bit error rate is disclosed. The memory array alternates odd data bits of a first plurality of data words in consecutive columns a first sub-bank of first and second memory banks and even data bits of the first plurality of data words in consecutive columns of a second sub-bank of the first and second memory banks. For example, the lowest bits of each of N data words are stored in a first N consecutive columns of a first sub-bank. The second bits of the N data words are stored in the next N consecutive columns of a second sub-bank. The N data bits in each of the bit positions of the N data words are interleaved in corresponding column mux sets. Alternating odd and even bits between sub-banks reduces multi-bit soft errors.