Abstract: Embodiments of systems, methods, and apparatuses for heterogeneous computing are described. In some embodiments, a hardware heterogeneous scheduler dispatches instructions for execution on one or more plurality of heterogeneous processing elements, the instructions corresponding to a code fragment to be processed by the one or more of the plurality of heterogeneous processing elements, wherein the instructions are native instructions to at least one of the one or more of the plurality of heterogeneous processing elements.

Abstract: Embodiments may be generally direct to apparatuses, systems, method, and techniques to detect a message to communicate via an interconnect coupled with a device capable of communication via a plurality of interconnect protocols, the plurality of interconnect protocols comprising a non-coherent interconnect protocol, a coherent interconnect protocol, and a memory interconnect protocol. Embodiments also include determining an interconnect protocol of the plurality of interconnect protocols to communicate the message via the interconnect based on the message, and providing the message to a multi-protocol multiplexer coupled with the interconnect, the multi-protocol multiplexer to communicate the message utilizing the interconnect protocol via the interconnect with the device.

Abstract: Aspects of the embodiments are directed to systems and methods for providing and using hints in data packets to perform memory transaction optimization processes prior to receiving one or more data packets that rely on memory transactions. The systems and methods can include receiving, from a device connected to the root complex across a PCIe-compliant link, a data packet; identifying from the received device a memory transaction hint bit; determining a memory transaction from the memory transaction hint bit; and performing an optimization process based, at least in part, on the determined memory transaction.

Abstract: Embodiments may be generally direct to apparatuses, systems, method, and techniques to provide multi-interconnect protocol communication. In an embodiment, an apparatus for providing multi-interconnect protocol communication may include a component comprising at least one connector operative to connect the component to at least one off-package device via a standard interconnect protocol, and logic, at least a portion of the logic comprised in hardware, the logic to determine data to be communicated via a multi-interconnect protocol, provide the data to a multi-protocol multiplexer to determine a route for the data, route the data on-package responsive to the multi-protocol multiplexer indicating a multi-interconnect on-package mode, and route the data off-package via the at least one connector responsive to the multi-protocol multiplexer indicating a multi-interconnect off-package mode. Other embodiments are described.

Abstract: Embodiments of systems, methods, and apparatuses for heterogeneous computing are described. In some embodiments, a hardware heterogeneous scheduler dispatches instructions for execution on one or more plurality of heterogeneous processing elements, the instructions corresponding to a code fragment to be processed by the one or more of the plurality of heterogeneous processing elements, wherein the instructions are native instructions to at least one of the one or more of the plurality of heterogeneous processing elements.

Abstract: A microelectronic assembly is provided comprising: a first plurality of integrated circuit (IC) dies arranged in an array of rows and columns in a first layer; and a second plurality of IC dies in a second layer not coplanar with the first layer. A first IC die in the first plurality is differently sized than surrounding IC dies in the first plurality, and a second IC die in the second plurality coupled to the first IC die comprises at least one of: a repeater circuitry and a fanout structure in an electrical pathway coupling the first IC die with an adjacent IC die in the first plurality.

Abstract: A fabric controller to provide a coherent accelerator fabric, including: a host interconnect to communicatively couple to a host device; a memory interconnect to communicatively couple to an accelerator memory; an accelerator interconnect to communicatively couple to an accelerator having a last-level cache (LLC); and an LLC controller configured to provide a bias check for memory access operations.

Abstract: Examples include techniques associated with mapping system memory physical addresses to proximity domains. Examples include mapping system memory physical addresses for a memory coupled with a multi-die system to proximity domains that include cores of a multi-core processor and the associated level 3 (L3) cache for use by each core included in a respective proximity domain. The mapping is to facilitate cache line ownership of a cache line in an L3 cache by an input/output device or agent located on a separate die from the multi-core processor.

Abstract: Systems and devices can include a controller and a command queue to buffer incoming write requests into the device. The controller can receive, from a client across a link, a non-posted write request (e.g., a deferred memory write (DMWr) request) in a transaction layer packet (TLP) to the command queue; determine that the command queue can accept the DMWr request; identify, from the TLP, a successful completion (SC) message that indicates that the DMWr request was accepted into the command queue; and transmit, to the client across the link, the SC message that indicates that the DMWr request was accepted into the command queue. The controller can receive a second DMWr request in a second TLP; determine that the command queue is full; and transmit a memory request retry status (MRS) message to be transmitted to the client in response to the command queue being full.

Abstract: Disclosed embodiments relate to remote atomic operations (RAO) in multi-socket systems. In one example, a method, performed by a cache control circuit of a requester socket, includes: receiving the RAO instruction from the requester CPU core, determining a home agent in a home socket for the addressed cache line, providing a request for ownership (RFO) of the addressed cache line to the home agent, waiting for the home agent to either invalidate and retrieve a latest copy of the addressed cache line from a cache, or to fetch the addressed cache line from memory, receiving an acknowledgement and the addressed cache line, executing the RAO instruction on the received cache line atomically, subsequently receiving multiple local RAO instructions to the addressed cache line from one or more requester CPU cores, and executing the multiple local RAO instructions on the received cache line independently of the home agent.

Abstract: Systems and devices can include a controller and a command queue to buffer incoming write requests into the device. The controller can receive, from a client across a link, a non-posted write request (e.g., a deferred memory write (DMWr) request) in a transaction layer packet (TLP) to the command queue; determine that the command queue can accept the DMWr request; identify, from the TLP, a successful completion (SC) message that indicates that the DMWr request was accepted into the command queue; and transmit, to the client across the link, the SC message that indicates that the DMWr request was accepted into the command queue. The controller can receive a second DMWr request in a second TLP; determine that the command queue is full; and transmit a memory request retry status (MRS) message to be transmitted to the client in response to the command queue being full.

Abstract: Aspects of the embodiments are directed to systems and methods for providing and using hints in data packets to perform memory transaction optimization processes prior to receiving one or more data packets that rely on memory transactions. The systems and methods can include receiving, from a device connected to the root complex across a PCIe-compliant link, a data packet; identifying from the received device a memory transaction hint bit; determining a memory transaction from the memory transaction hint bit; and performing an optimization process based, at least in part, on the determined memory transaction.

Abstract: Disclosed embodiments relate to spatial and temporal merging of remote atomic operations.

Abstract: Embodiments of systems, methods, and apparatuses for heterogeneous computing are described. In some embodiments, a hardware heterogeneous scheduler dispatches instructions for execution on one or more plurality of heterogeneous processing elements, the instructions corresponding to a code fragment to be processed by the one or more of the plurality of heterogeneous processing elements, wherein the instructions are native instructions to at least one of the one or more of the plurality of heterogeneous processing elements.

Abstract: Embodiments may be generally direct to apparatuses, systems, method, and techniques to detect a message to communicate via an interconnect coupled with a device capable of communication via a plurality of interconnect protocols, the plurality of interconnect protocols comprising a non-coherent interconnect protocol, a coherent interconnect protocol, and a memory interconnect protocol. Embodiments also include determining an interconnect protocol of the plurality of interconnect protocols to communicate the message via the interconnect based on the message, and providing the message to a multi-protocol multiplexer coupled with the interconnect, the multi-protocol multiplexer to communicate the message utilizing the interconnect protocol via the interconnect with the device.

Abstract: Aspects of the embodiments are directed to systems and methods for providing and using hints in data packets to perform memory transaction optimization processes prior to receiving one or more data packets that rely on memory transactions. The systems and methods can include receiving, from a device connected to the root complex across a PCIe-compliant link, a data packet; identifying from the received device a memory transaction hint bit; determining a memory transaction from the memory transaction hint bit; and performing an optimization process based, at least in part, on the determined memory transaction.

Abstract: Embodiments of systems, methods, and apparatuses for heterogeneous computing are described. In some embodiments, a hardware heterogeneous scheduler dispatches instructions for execution on one or more plurality of heterogeneous processing elements, the instructions corresponding to a code fragment to be processed by the one or more of the plurality of heterogeneous processing elements, wherein the instructions are native instructions to at least one of the one or more of the plurality of heterogeneous processing elements.

Abstract: A fabric controller to provide a coherent accelerator fabric, including: a host interconnect to communicatively couple to a host device; a memory interconnect to communicatively couple to an accelerator memory; an accelerator interconnect to communicatively couple to an accelerator having a last-level cache (LLC); and an LLC controller configured to provide a bias check for memory access operations.

Abstract: Disclosed embodiments relate to remote atomic operations (RAO) in multi-socket systems. In one example, a method, performed by a cache control circuit of a requester socket, includes: receiving the RAO instruction from the requester CPU core, determining a home agent in a home socket for the addressed cache line, providing a request for ownership (RFO) of the addressed cache line to the home agent, waiting for the home agent to either invalidate and retrieve a latest copy of the addressed cache line from a cache, or to fetch the addressed cache line from memory, receiving an acknowledgement and the addressed cache line, executing the RAO instruction on the received cache line atomically, subsequently receiving multiple local RAO instructions to the addressed cache line from one or more requester CPU cores, and executing the multiple local RAO instructions on the received cache line independently of the home agent.

Abstract: A fabric controller is provided for a coherent accelerator fabric. The coherent accelerator fabric includes a host interconnect, a memory interconnect, and an accelerator interconnect. The host interconnect communicatively couples to a host device. The memory interconnect communicatively couples to an accelerator memory. The accelerator interconnect communicatively couples to an accelerator having a last-level cache (LLC). An LLC controller is provided that is configured to provide a bias check for memory access operations on the fabric.