Abstract: Instructions and logic provide for a Single Instruction Multiple Data (SIMD) SM4 round slice operation. Embodiments of an instruction specify a first and a second source data operand set, and substitution function indicators, e.g. in an immediate operand. Embodiments of a processor may include encryption units, responsive to the first instruction, to: perform a slice of SM4-round exchanges on a portion of the first source data operand set with a corresponding keys from the second source data operand set in response to a substitution function indicator that indicates a first substitution function, perform a slice of SM4 key generations using another portion of the first source data operand set with corresponding constants from the second source data operand set in response to a substitution function indicator that indicates a second substitution function, and store a set of result elements of the first instruction in a SIMD destination register.

Abstract: A photoresist is disclosed. The photoresist includes a polymer with one repeating unit and an absorbing unit.

Abstract: Methods, apparatus, systems and articles of manufacture are disclosed that enable out-of-order pipelined execution of static mapping of a workload to one or more computational building blocks of an accelerator. An example apparatus includes an interface to load a first number of credits into memory; a comparator to compare the first number of credits to a threshold number of credits associated with memory availability in a buffer; and a dispatcher to, when the first number of credits meets the threshold number of credits, select a workload node of the workload to be executed at a first one of the one or more computational building blocks.

Abstract: An apparatus to facilitate security within a computing system is disclosed. The apparatus includes a storage drive, a controller, comprising a trusted port having one or more key slots to program one or more cryptographic keys and an encryption engine to receive the cryptographic keys via the one or more key slots, encrypt data written to the storage drive using the cryptographic keys and decrypt data read from the storage drive using the cryptographic keys.

Abstract: Embodiments detailed herein relate to matrix (tile) operations. For example, decode circuitry to decode an instruction having fields for an opcode and a memory address; and execution circuitry to execute the decoded instruction to set a tile configuration for the processor to utilize tiles in matrix operations based on a description retrieved from the memory address, wherein a tile a set of 2-dimensional registers are discussed.

Abstract: Apparatus and methods for bone conduction detection are disclosed herein. An example apparatus includes memory; machine-readable instructions; and processor circuitry to execute the machine-readable instructions to associate a vibration signal with a voice or with motion, the vibration signal transmitted via a bone structure of a user; permit access to a user application based on the association of the vibration signal with the voice; and deny access to the user application based on the association of the vibration signal with the motion.

Abstract: Disclosed herein are IC structures, packages, and devices that include transistors, e.g., III-N transistors, having a source region, a drain region (together referred to as “source/drain” (S/D) regions), and a gate stack. In one aspect, a contact to at least one of the S/D regions of a transistor may have a width that is smaller than a width of the S/D region. In another aspect, a contact to a gate electrode material of the gate stack of a transistor may have a width that is smaller than a width of the gate electrode material. Reducing the width of contacts to S/D regions or gate electrode materials of a transistor may reduce the overlap area between various pairs of these contacts, which may, in turn, allow reducing the off-state capacitance of the transistor. Reducing the off-state capacitance of III-N transistors may advantageously allow increasing their switching frequency.

Abstract: Methods, apparatuses, and computer readable media for report identification and power control for multi-device wireless sensing in a wireless network are disclosed. An apparatus of an access point (AP) is disclosed, where the apparatus comprises processing circuitry configured to encode for transmission to a station (STA) a frame comprising a configuration hold subfield, the hold subfield indicating whether the STA should maintain a transmit configuration unchanged during uplink (UL) sensing. The processing circuitry further configured to encode for transmission a trigger frame (TF) for sensing, the TF for sensing including an identification of the STA, an indication of a resource unit (RU) for the STA to use to transmit an uplink (UL) sensing packet, and TF type subfield, the a TF type subfield indicating the TF for sensing.

Abstract: Systems, methods, and apparatuses relating to circuitry to implement a duplication resistant on-die irregular data prefetcher are described.

Abstract: Techniques are provided for processing video frames in a process flow that includes first and second computation engines. In an example, the first engine is an artificial intelligence based computation engine, and the second engine is a heuristics-based computation engine. A sequence of frames of a video includes a first and second frames that are two consecutive frames in the sequence. An analyzer determines whether the second frame has non-redundant information relative to the first frame. In response to the determination, the analyzer selects one of the first or second engine for processing at least a section of the second frame. For example, if the second frame has non-redundant information relative to the first frame, at least the section of the second frame is processed by the first engine. If the second frame does not include non-redundant information, the second frame is processed by the second engine.

Abstract: An apparatus and method are described for managing data which is biased towards a processor or a GPU. For example, an apparatus comprises a processor comprising one or more cores, one or more cache levels, and cache coherence controllers to maintain coherent data in the one or more cache levels; a graphics processing unit (GPU) to execute graphics instructions and process graphics data, wherein the GPU and processor cores are to share a virtual address space for accessing a system memory; a GPU memory addressable through the virtual address space shared by the processor cores and GPU; and bias management circuitry to store an indication for whether the data has a processor bias or a GPU bias, wherein if the data has a GPU bias, the data is to be accessed by the GPU without necessarily accessing the processor's cache coherence controllers.

Abstract: A microelectronic package may be fabricated having a microelectronic die stack attached to a microelectronic substrate and at least one microelectronic device, which is separate from the microelectronic die stack, attached to the microelectronic substrate within the footprint of one of the microelectronic dice within the microelectronic die stack. In one embodiment, the microelectronic die stack may have a plurality of stacked microelectronic dice, wherein one microelectronic die of the plurality of microelectronic dice has a footprint greater than the other microelectronic die of the plurality of microelectronic dice, and wherein the at least one microelectronic device is attached to the one microelectronic die of the plurality of microelectronic dice having the greater footprint.

Abstract: Systems, methods, and apparatuses relating to instructions to multiply values of zero are described.

Abstract: Techniques for embedding fabric addressing information within Ethernet media access control (MAC) addresses is disclosed herein and allows a multi-node fabric having potentially millions of nodes to feature Ethernet encapsulation without the necessity of a lookup or map to translate MAC addresses to fabric-routable local identifiers (LIDs). In particular, a locally-administered MAC address may be encoded with fabric addressing information including a LID. Thus a node may exchange Ethernet packets using a multi-node fabric by encapsulating each Ethernet packet with a destination MAC address corresponding to an intended destination. As the destination MAC address may implicitly map to a LID of the multi-node fabric, the node may use an extracted LID value therefrom to address a fabric-routable packet. To this end, a node may introduce a fabric-routable packet encapsulating an Ethernet packet onto a multi-node fabric without necessarily performing a lookup to map a MAC address to a corresponding LID.

Abstract: Systems for providing prioritization of UL transmissions in a UE are described. The prioritization information is used to resolve resource conflicts among UL transmissions that include conflicts between high priority UL transmissions, between an aperiodic-channel state information transmission and a scheduling request, and between a low priority UL transmission and a high priority UL transmission when timeline conditions for multiplexing in a single UL transmission are not met. The prioritization is based on timing and priority of the UL transmissions to determine which of the UL transmissions to transmit and which to cancel. Additional prioritization is based on reception by the UE of a cancelation index or in an additional overlapping high priority UL grant received in a DCI of a PDCCH that overlaps with at least one other PDCCH associated with the UL transmissions.

Abstract: A generation-Node B (gNB) configured for unlicensed spectrum operation above 52.6 GHz in a fifth-generation new-radio (NR) system (5GS) may encode a parameter (e.g., ssb-PositionsInBurst) for transmission to a UE (e.g., in the SIM or UE specific RRC signalling). The parameter may indicate candidate positions of synchronization signal blocks (SSBs) within a discovery reference signal (DRS) measurement timing configuration (DMTC) transmission window within slots of a system frame (SFN). During the DMTC window, the gNB may perform a LBT procedure on an unlicensed carrier of the unlicensed spectrum to determine if the unlicensed carrier is available. When the LBT is successful (i.e., the unlicensed carrier is available), the gNB may encode a discovery reference signal (DRS) for transmission on the unlicensed carrier. The DRS may include one or more of the SSBs transmitted during the candidate positions that fall within the DRS.

Abstract: Systems and methods include establishing a cryptographically secure communication between an application module and an audio module. The application module is configured to execute on an information-handling machine, and the audio module is coupled to the information-handling machine. The establishment of the cryptographically secure communication may be at least partially facilitated by a mutually trusted module.

Abstract: A platform comprising numerous reconfigurable circuit components arranged to operate as primary and redundant circuits is provided. The platform further comprises security circuitry arranged to monitor the primary circuit for anomalies and reconfigurable circuit arranged to disconnect the primary circuit from a bus responsive to detection of an anomaly. Furthermore, the present disclosure provides for the quarantine, refurbishment and designation as redundant, the anomalous circuit.

Abstract: Disclosed embodiments relate to accelerating multiplication of sparse matrices. In one example, a processor is to fetch and decode an instruction having fields to specify locations of first, second, and third matrices, and an opcode indicating the processor is to multiply and accumulate matching non-zero (NZ) elements of the first and second matrices with corresponding elements of the third matrix, and executing the decoded instruction as per the opcode to generate NZ bitmasks for the first and second matrices, broadcast up to two NZ elements at a time from each row of the first matrix and each column of the second matrix to a processing engine (PE) grid, each PE to multiply and accumulate matching NZ elements of the first and second matrices with corresponding elements of the third matrix. Each PE further to store an NZ element for use in a subsequent multiplications.

Abstract: Alternative surfaces for conductive pad layers of silicon bridges for semiconductor packages, and the resulting silicon bridges and semiconductor packages, are described. In an example, a semiconductor structure includes a substrate having a lower insulating layer disposed thereon. The substrate has a perimeter. A metallization structure is disposed on the lower insulating layer. The metallization structure includes conductive routing disposed in a dielectric material stack. First and second pluralities of conductive pads are disposed in a plane above the metallization structure. Conductive routing of the metallization structure electrically connects the first plurality of conductive pads with the second plurality of conductive pads. An upper insulating layer is disposed on the first and second pluralities of conductive pads. The upper insulating layer has a perimeter substantially the same as the perimeter of the substrate.

Abstract: Apparatuses, methods and storage medium associated with embedded computing, are disclosed herein. In embodiments, an embedded computing platform includes a plurality of system-on-chips (SoCs) forming a local compute cluster; and an orchestrator disposed on one of the SoCs arranged to orchestrate fail-safe operations, in response to a reported unrecoverable failure requiring shut down or partial disabling of one of the SoCs, to consolidate execution of critical workloads on one or more of remaining fully or partially operational ones of the SoCs. Other embodiments are also described and claimed.

Abstract: Embodiments disclosed herein include three-dimensional 3D arrays of memory cells and methods of forming such devices. In an embodiment a memory device comprises, a substrate surface, and a three-dimensional (3D) array of memory cells over the substrate surface. In an embodiment each memory cell comprises a transistor and a capacitor. In an embodiment the transistor of each memory cell comprises, a semiconductor channel, with a first end of the semiconductor channel electrically coupled to a bit line that runs substantially parallel to the substrate surface, and a second end of the semiconductor channel is electrically coupled to the capacitor. The transistor may also comprise a gate dielectric on a surface of the semiconductor channel between the first end and the second end of the semiconductor channel. In an embodiment, the gate dielectric is contacted by a word line that runs substantially perpendicular to the substrate surface.

Abstract: Systems and methods are disclosed for providing non-lexical cues in synthesized speech. An example system includes processor circuitry to generate a breathing cue to enhance speech to be synthesized from text; determine a first insertion point of the breathing cue in the text, wherein the breathing cue is identified by a first tag of a markup language; generate a prosody cue to enhance speech to be synthesized from the text; determine a second insertion point of the prosody cue in the text, wherein the prosody cue is identified by a second tag of the markup language; insert the breathing cue at the first insertion point based on the first tag and the prosody cue at the second insertion point based on the second tag; and trigger a synthesis of the speech from the text, the breathing cue, and the prosody cue.

Abstract: Embodiments described herein provide techniques for forming an interconnect structure that includes micro features formed therein. Such embodiments can assist with improving interconnect joint reliability when compared to conventional pads that have a flat surface. An interconnect structure may comprise: a metal pad over a substrate (e.g., a semiconductor package, a PCB, an interposer, etc.). Micro features may be formed in an edge of the metal pad or away from the edge of the metal pad. The micro features can assist with: (i) increasing the contact area between solder used to form an interconnect joint and the metal pad; and (ii) improving adherence of solder used to form an interconnect joint to the metal pad. These benefits can improve interconnect joint reliability by, among others, improving the interconnect joint's ability to absorb stress from substrates having differing coefficients of thermal expansion.

Abstract: Interconnect via metal-insulator-metal (MIM) fuse for integrated circuitry. Two electrode metallization features, which may be within a backend of an IC die, are coupled through a via comprising a fuse material layer. The fuse material layer passes a non-zero leakage current when a lower read voltage is applied across the electrode metallization features, and irreversibly forms an open circuit when a higher programming voltage is applied across the electrode metallization features. The fuse material layer may be a compound of a metal and oxygen and be sufficiently thin to ensure a significant leakage current at the read voltage. Joule heating of the fuse material layer may induce a void between the electrode metallization features as the leakage current through the fuse material layer increases under higher voltages, creating an open circuit.

Abstract: One embodiment provides a general-purpose graphics processing unit comprising a dynamic precision floating-point unit including a control unit having precision tracking hardware logic to track an available number of bits of precision for computed data relative to a target precision, wherein the dynamic precision floating-point unit includes computational logic to output data at multiple precisions.

Abstract: An integrated circuit structure includes a second device stacked vertically above a first device. The first device includes (i) a first source or drain region, (ii) a first source or drain contact coupled to the first source or drain region, and (iii) a first layer comprising a first metal and first one or more semiconductor materials between at least a section of the first source or drain region and the first source or drain contact. The second device includes (i) a second source or drain region, (ii) a second source or drain contact coupled to the second source or drain region, and (iii) a second layer comprising a second metal and second one or more semiconductor materials between at least a section of the second source or drain region and the second source or drain contact. In an example, the first metal and the second metal are different.

Abstract: Described herein is a generic hardware/software communication (HSC) channel that facilitates the re-use of pre-silicon DPI methods to enable FPGA-based post-silicon validation. The HSC channel translates a DPI interface into a hardware FIFO based mechanism. This translation allows the reuse of the methods without having to re-implement the entire flow in pure hardware. The core logic for the transactor remains the same, while only a small layer of the transactor is converted into the FIFO based mechanism.

Abstract: The technology described herein includes a first plurality of bijection diffusion function circuits to diffuse data bits into diffused data bits and store the diffused data bits into a memory; an error correcting code (ECC) generation circuit to generate ECC bits for the data bits; and a second plurality of bijection diffusion function circuits to diffuse the ECC bits into diffused ECC bits and store the diffused ECC bits into the memory.

Abstract: Disclosed herein are integrated circuit (IC) structures including backside vias, as well as related methods and devices. In some embodiments, an IC structure may include: a device layer, wherein the device layer includes a plurality of active devices; a first metallization layer over the device layer, wherein the first metallization layer includes a first conductive pathway in conductive contact with at least one of the active devices in the device layer; a second metallization layer under the device layer, wherein the second metallization layer includes a second conductive pathway; and a conductive via in the device layer, wherein the conductive via is in conductive contact with at least one of the active devices in the device layer and also in conductive contact with the second conductive pathway.

Abstract: An apparatus to facilitate encoding video data is disclosed. The apparatus includes rendering logic to render graphics video data as frame data, fade extractor logic to extract fade effects data to be applied to the frame data to generate frame auxiliary metadata comprising the fade effects data, weighted prediction logic to receive the frame data and the auxiliary metadata and compute one or more weighted predictions on the frame data at one or more frame sequences indicated in the fade effects data and encoding logic to encode the frame data based on the one or more weighted predictions.

Abstract: Techniques for thin-film resistors in vias are disclosed. In the illustrative embodiment, thin-film resistors are formed in through-glass vias of a glass substrate of an interposer. The thin-film resistors do not take up a significant amount of area on a layer of the interposer, as the thin-film resistor extends vertically through a via rather than horizontally on a layer of the interposer. The thin-film resistors may be used for any suitable purpose, such as power dissipation or voltage control, current control, as a pull-up or pull-down resistor, etc.

Abstract: A graphics processing device is provided that includes a set of compute units to execute a workload, a cache coupled with the set of compute units, and circuitry coupled with the cache and the set of compute units. The circuitry is configured to, in response to a cache miss for the read from a first cache, broadcast an event within the graphics processor device to identify data associated with the cache miss, receive the event at a second compute unit in the set of compute units, and prefetch the data identified by the event into a second cache that is local to the second compute unit before an attempt to read the instruction or data by the second thread.

Abstract: Capacitors for decoupling, power delivery, integrated circuits, related systems, and methods of fabrication are disclosed. Such capacitors include a transition metal oxide dielectric between two electrodes, at least one of which includes a conductive metal oxide layer on the transition metal oxide dielectric and a high density metal layer on the conductive metal oxide.

Abstract: One embodiment provides an apparatus comprising a circuit board; an active interposer coupled with the circuit board via a debug package, and a graphics processor die coupled with the active interposer via the debug package. The graphics processor die includes graphics processor resources configured to execute instructions for a multi-die system on chip (SoC) device and excludes functionality that is implemented in a separate due of the multi-die SoC. The apparatus includes a field-programmable gate array (FPGA) including hardware logic that is configurable to emulate functionality provided by the separate die of the multi-die SoC device, which enables silicon validation of the graphics processor die separately from other dies of the multi-die SoC.

Abstract: Clock distribution in an integrated circuit component can comprise the generation of bulk acoustic waves by acoustic transmitters and propagation of the bulk acoustic waves across the substrate where they are received by piezoelectric elements acting as acoustic receivers. Clock distribution can also comprise the generation of surface acoustic waves by acoustic transmitters located on the same substrate surface as the piezoelectric elements. An acoustic transmitter comprises a layer of piezoelectric material that generates an acoustic wave in response to the piezoelectric layer being activated by a clock source signal applied to the acoustic transmitter. The piezoelectric elements convert the acoustic waves into an electrical signal which can be used as a local clock signal for devices and components in the vicinity of the piezoelectric elements or from which such a local clock signal can be derived.

Abstract: An integrated circuit structure includes a device including a source region, a drain region, a body laterally between the source and drain regions, and a source contact coupled to the source region. In an example, the source region includes a first region, and a second region compositionally different from and above the first region. The source contact extends through the second region and extends within the first region. In an example where the device is a p-channel metal-oxide-semiconductor (PMOS) device, a concentration of germanium within the second region is different (e.g., higher) than a concentration of germanium within the first region. In another example where the device is a n-channel metal-oxide-semiconductor (NMOS) device, a doping concentration level of a dopant (e.g., an n-type dopant) within the second region is different (e.g., higher) from a doping concentration level of the dopant within the first region.

Abstract: Various embodiments are generally directed to techniques for memory access by a computer in a reduced power state, such as during video playback or connected standby. Some embodiments are particularly directed to disabling one or more memory channels during a reduced power state by mapping memory usages during the reduced power state to one of a plurality of memory channels. In one embodiment, for example, one or more low-power mode blocks in a set of functional blocks of a computer may be identified. In some such embodiments, the computer may include a processor, a memory, and first and second memory channels to communicatively couple the processor with the second memory. In many embodiments, usage of the one or more low-power mode blocks in the set of functional blocks may be mapped to a first address range associated with the first memory channel.

Abstract: An apparatus is described. The apparatus includes a long short term memory (LSTM) circuit having a multiply accumulate circuit (MAC). The MAC circuit has circuitry to rely on a stored product term rather than explicitly perform a multiplication operation to determine the product term if an accumulation of differences between consecutive, preceding input values has not reached a threshold.

Abstract: Deep neural networks (DNNs) with budding ensemble architectures may be trained using diversity loss. A DNN may include a backbone and a plurality of heads. The backbone includes one or more layers. A layer in the backbone may generate an intermediate tensor. The plurality of heads may include one or more pairs of heads. A pair of heads includes a first head and a second head duplicated from the first head. The second head may include the same tensor operations as the first head but different internal parameters. The intermediate tensor generated by a backbone layer may be input into both the first head and the second head. The first head may compute a first detection tensor, and the second head may compute a second detection tensor. A similarity between the first detection tensor and the second detection tensor may be used as a diversity loss for training the DNN.

Abstract: An apparatus and method are provided which take advantage of heterogeneous compute capability to dynamically pick the best operating core for BIOS power-up flows and sleep exit flows (e.g., S3, S4, and/or S5). The selection of the BSP is moved to an early power-up time instead of a fixed hardware selection at any time. For maximum boot performance, the system selects the fastest capable core as the BSP at an early power-up time. In addition, for maximum power saving, the system selects the most power efficient core as the BSP. Processor or switching for selecting the BSP happens during the boot-up as well as power-up flows (e.g., S3, S4, and/or S5 flows).

Abstract: Technologies for processing network packets by a host interface of a network interface controller (NIC) of a compute device. The host interface is configured to retrieve, by a symmetric multi-purpose (SMP) array of the host interface, a message from a message queue of the host interface and process, by a processor core of a plurality of processor cores of the SMP array, the message to identify a long-latency operation to be performed on at least a portion of a network packet associated with the message. The host interface is further configured to generate another message which includes an indication of the identified long-latency operation and a next step to be performed upon completion. Additionally, the host interface is configured to transmit the other message to a corresponding hardware unit scheduler as a function of the subsequent long-latency operation to be performed. Other embodiments are described herein.

Abstract: An integrated circuit (IC) assembly comprising an IC die and a frame material that has been dispensed over the assembly substrate to be further adjacent to a perimeter edge of the IC die. The frame material may be selected to have flow properties that minimize slump, for example so a profile of a transverse cross-section through the frame material may retain convex curvature. The frame material may be cured following dispense, and upon application of a thermal interface material (TIM), the frame material may and act as a barrier, impeding flow of the TIM. The frame material may be compressed by force applied through an external thermal solution, such as a heat sink, to ensure good contact to the TIM.

Abstract: A Bluetooth receiver is provided. The Bluetooth receiver comprises interface circuitry configured to receive a receive packet. Further, the Bluetooth receiver comprises physical layer processing circuitry configured to demodulate the receive packet into a bit stream representing a sequence of data symbols. Further, the physical layer configured to determine a number of bits in the bit stream having a highest likelihood of being erroneous as weak-bits and determine locations of the identified weak-bits in the bit stream. The Bluetooth receiver further comprises medium access control layer processing circuitry configured to receive the bit stream and information about the determined locations of the identified weak-bits from the physical layer processing circuitry.

Abstract: Embodiments herein describe techniques for a semiconductor device including a transistor. The transistor includes a first metal contact as a source electrode, a second metal contact as a drain electrode, a channel area between the source electrode and the drain electrode, and a third metal contact aligned with the channel area as a gate electrode. The first metal contact may be located in a first metal layer along a first direction. The second metal contact may be located in a second metal layer along the first direction, in parallel with the first metal contact. The third metal contact may be located in a third metal layer along a second direction substantially orthogonal to the first direction. The third metal layer is between the first metal layer and the second metal layer. Other embodiments may be described and/or claimed.

Abstract: Methods and devices including a processing circuit for a User Equipment, UE, wherein the UE includes a radio frequency, RF, receiver, wherein the processing circuit is configured to set a power switching pattern of the RF receiver for UE power saving in Discontinuous Reception, DRX, Radio Resource Control, RRC, IDLE mode based on a paging occasion, PO, and a signal block associated with the PO.

Abstract: Embodiments described herein include software, firmware, and hardware logic that provides techniques to perform arithmetic on sparse data via a systolic processing unit. One embodiment provides for data aware sparsity via compressed bitstreams. One embodiment provides for block sparse dot product instructions. One embodiment provides for a depth-wise adapter for a systolic array.

Abstract: In one embodiment, a multi-tenant computing system includes at least one processor including a plurality of cores on which a plurality of agents of a plurality of tenants of the multi-tenant computing system are to execute, a configuration storage, and a memory execution circuit. The configuration storage includes a first configuration register to store configuration information associated with the memory execution circuit. The first configuration register is to store a mode identifier to identify a mode of operation of the memory execution circuit. The memory execution circuit, in a first mode of operation, is to receive encrypted data of a first tenant of the plurality of tenants, the encrypted data encrypted by the first tenant, generate an integrity value for the encrypted data, and send the encrypted data and the integrity value to a memory, wherein the integrity value is not visible to the software of the multi-tenant computing system.

Abstract: Techniques related to video coding include multiple channel video coding with cross-channel referencing.

Abstract: Examples relate to a die interconnect substrate comprising a bridge die comprising at least one bridge interconnect connecting a first bridge die pad of the bridge die to a second bridge die pad of the bridge die. The die interconnect substrate further comprises a substrate structure comprising a substrate interconnect electrically insulated from the bridge die, wherein the bridge die is embedded in the substrate structure. The die interconnect substrate further comprises a first interface structure for attaching a semiconductor die to the substrate structure, wherein the first interface structure is connected to the first bridge die pad. The die interconnect substrate further comprises a second interface structure for attaching a semiconductor die to the substrate structure, wherein the second interface structure is connected to the substrate interconnect. A surface of the first interface structure and a surface of the second interface structure are at the same height.