Abstract: Structures having magnetic vias and backside power delivery are described. In an example, an integrated circuit structure includes a front-side structure including a device layer having a plurality of nanowire-based transistors, and a plurality of metallization layers above the nanowire-based or fin-based transistors of the device layer. One of the metallization layers includes one or more magnetic vias. A backside structure is below the nanowire-based or fin-based transistors of the device layer. The backside structure includes a ground metal line.

Abstract: Described is an low overhead method and apparatus to reconfigure a pair of buffered interconnect links to operate in one of these three modes—first mode (e.g., bandwidth mode), second mode (e.g., latency mode), and third mode (e.g., energy mode). In bandwidth mode, each link in the pair buffered interconnect links carries a unique signal from source to destination. In latency mode, both links in the pair carry the same signal from source to destination, where one link in the pair is “primary” and other is called the “assist”. Temporal alignment of transitions in this pair of buffered interconnects reduces the effective capacitance of primary, thereby reducing delay or latency. In energy mode, one link in the pair, the primary, alone carries a signal, while the other link in the pair is idle. An idle neighbor on one side reduces energy consumption of the primary.

Abstract: Various embodiments provide apparatuses, systems, and methods for resonant rotary clocking to generate synchronized clock signals. A base die may include a resonant ring structure to form a plurality of rotary traveling wave oscillators (RTWOs) coupled to one another in a rotary oscillator array (ROA). The ROA may provide synchronized clock signals at deterministic phase points that are tapped from the resonant ring structure. Multiple dies may be coupled to the base die (e.g., in a multi-die system) and may receive the tapped clock signals. Other embodiments may be described and claimed.

Abstract: A microelectronic assembly is provided comprising: a first plurality of integrated circuit (IC) dies in a first layer; a second plurality of IC dies in a second layer between the first layer and a third layer; and a third plurality of IC dies in the third layer. In some embodiments, the second plurality of IC dies comprises IC dies in an array of rows and columns, each IC die of the second plurality of IC dies is coupled to more than one IC die of the first plurality of IC dies, and the third plurality of IC dies is to provide electrical coupling between adjacent ones of the second plurality of IC dies.

Abstract: A hash accelerator of a cache memory may receive a query from a processor comprising the cache memory, the query to comprise an input key and an operation to be performed based on a hash table stored in the cache memory. The hash accelerator may determine whether an entry associated with the input key exists in a lock board of the hash accelerator. The hash accelerator may process the query based on whether the entry exists in the lock board.

Abstract: Embodiments disclosed herein include dies and die modules. In an embodiment, a die comprises a substrate with a first surface and a second surface opposite from the first surface. In an embodiment the substrate comprises a semiconductor material. In an embodiment, first bumps with a first pitch are on the first surface of the substrate. In an embodiment, a first layer surrounds the first bumps, where the first layer comprises a dielectric material. In an embodiment, second bumps with a second pitch are on the substrate. In an embodiment, the second pitch is greater than the first pitch. In an embodiment, a second layer surrounds the second bumps, where the second layer comprises a dielectric material.

Abstract: Embodiments disclosed herein include electronic packages. In an embodiment, an electronic package comprises a package substrate, and a die module coupled to the package substrate. In an embodiment, the die module comprises a die and a chiplet coupled to the die. In an embodiment, the chiplet is coupled to the die with a hybrid bonding interconnect architecture.

Abstract: Embodiments disclosed herein include die modules. In an embodiment, a die module comprises a plurality of first dies, and a second die under the plurality of first dies. In an embodiment, the second die is coupled to individual ones of the plurality of first dies. In an embodiment, the second die comprises a plurality of mesh stops, and conductive routing to electrically couple the mesh stops together.

Abstract: Methods, apparatus, systems, and articles of manufacture providing a Bayesian compute unit with reconfigurable sampler and methods and apparatus to operate the same are disclosed. An example apparatus includes a number generator to generate a sequence of numbers; a multiplier to generate a plurality of products by multiplying respective numbers of the sequence of the numbers by a variance value; and an adder to generate a plurality of weights by adding a mean value to the plurality of products, the plurality of weights corresponding to a single probability distribution.

Abstract: An apparatus is provided which comprises a full adder including magnetoelectric material and spin orbit material. In some embodiments, the adder includes: a 3-bit carry generation structure and a multi-bit sum generation structure coupled to the 3-bit carry generation structure. In some embodiments, the 3-bit carry generation structure includes at least three cells comprising magnetoelectric material and spin orbit material, wherein the 3-bit carry generation structure is to perform a minority logic operation on first, second, and third inputs to generate a carry output. In some embodiments, the multi-bit sum generation structure includes at least four cells comprising magnetoelectric material and spin orbit material, wherein the multi-bit sum generation structure is to perform a minority logic operation on the first, second, and third inputs and the carry output to generate a sum output.

Abstract: An apparatus is provided which comprises one or more magnetoelectric spin orbit (MESO) minority gates with different peripheral complementary metal oxide semiconductor (CMOS) circuit techniques in the device layer including: (1) current mirroring, (2) complementary supply voltages, (3) asymmetrical transistor sizing, and (4) using transmission gates. These MESO minority gates use the multi-phase clock to prevent back propagation of current so that MESO gate can correctly process the input data.

Abstract: A DNN accelerator includes a multiplication controller controlling whether to perform matrix computation based on weight values. The multiplication controller reads a weight matrix from a WRAM in the DNN accelerator and determines a row value for a row in the weight matrix. In an embodiment where the row value is one, a first switch sends a read request to the WRAM to read weights in the row and a second switch forms a data transmission path from an IRAM in the DNN accelerator to a PE in the DNN accelerator. The PE receives the weights and input data stored in the IRAM and performs MAC operations. In an embodiment where the row value is zero, the first and second switches are not triggered. No read request is sent to the WRAM and the data transmission path is not formed. The PE will not perform any MAC operations.

Abstract: An apparatus is provided which comprises: a first Power Management Unit (PMU); and a second PMU, wherein the first PMU is to manage transition of the apparatus from a low power state to a first active state, wherein the second PMU is to manage transition of the apparatus from the first active state to a second active state, and wherein the second PMU is to be powered down while the apparatus is to be in the low power state.

Abstract: Methods, apparatus, systems, and articles of manufacture providing a Bayesian compute unit with reconfigurable sampler and methods and apparatus to operate the same are disclosed. An example apparatus includes a processor element to generate (a) a first element by applying a mean value to an activation and (b) a second element by applying a variance value to a square of the activation, the mean value and the variance value corresponding to a single probability distribution; a programmable sampling unit to: generate a pseudo random number; and generate an output based on the pseudo random number, the first element, and the second element, wherein the output corresponds to the single probability distribution; and output memory to store the output.

Abstract: Described are mechanisms and methods for amortizing the cost of address decode, row-decode and wordline firing across multiple read accesses (instead of just on one read access). Some or all memory locations that share a wordline (WL) may be read, by walking through column multiplexor addresses (instead of just reading out one column multiplexor address per WL fire or memory access). The mechanisms and methods disclosed herein may advantageously enable N distinct memory words to be read out if the array uses an N-to-1 column multiplexor. Since memories such as embedded DRAMs (eDRAMs) may undergo a destructive read, for a given WL fire, a design may be disposed to sense N distinct memory words and restore them in order.

Abstract: Described is an apparatus to reduce or eliminate imprint charge, wherein the apparatus which comprises: a source line; a bit-line; a memory bit-cell coupled to the source line and the bit-line; a first multiplexer coupled to the bit-line; a second multiplexer coupled to the source-line; a first driver coupled to the first multiplexer; a second driver coupled to the second multiplexer; and a current source coupled to the first and second drivers.

Abstract: Described is an low overhead method and apparatus to reconfigure a pair of buffered interconnect links to operate in one of these three modes—first mode (e.g., bandwidth mode), second mode (e.g., latency mode), and third mode (e.g., energy mode). In bandwidth mode, each link in the pair buffered interconnect links carries a unique signal from source to destination. In latency mode, both links in the pair carry the same signal from source to destination, where one link in the pair is “primary” and other is called the “assist”. Temporal alignment of transitions in this pair of buffered interconnects reduces the effective capacitance of primary, thereby reducing delay or latency. In energy mode, one link in the pair, the primary, alone carries a signal, while the other link in the pair is idle. An idle neighbor on one side reduces energy consumption of the primary.

Abstract: Systems, apparatuses and methods may provide for early pre-charge with respect to peak power events. Application performance may improve by pre-charging a supercap just prior to initiating a system wake up from a qualified system wake-source trigger. Additionally, the pre-charging of the supercap may be controlled by a time defined pre-charge period and may also be controlled by a predetermined threshold voltage.

Abstract: Described is an apparatus which comprises: a first electrical path comprising at least one driver and receiver; and a second electrical path comprising at least one driver and receiver, wherein the first and second electrical paths are to receive a same input signal, wherein the first electrical path and the second electrical path are parallel to one another and have substantially the same propagation delays, and wherein the second electrical path is enabled during a first operation mode and disabled during a second operation mode.

Abstract: Described are mechanisms and methods for amortizing the cost of address decode, row-decode and wordline firing across multiple read accesses (instead of just on one read access). Some or all memory locations that share a wordline (WL) may be read, by walking through column multiplexor addresses (instead of just reading out one column multiplexor address per WL fire or memory access). The mechanisms and methods disclosed herein may advantageously enable N distinct memory words to be read out if the array uses an N-to-1 column multiplexor. Since memories such as embedded DRAMs (eDRAMs) may undergo a destructive read, for a given WL fire, a design may be disposed to sense N distinct memory words and restore them in order.