Abstract: An embodiment of an integrated circuit comprises circuitry to share page tables associated with a page between a processor memory management unit (MMU) and an input/output memory management unit (IOMMU), store a page table entry in the memory associated with the page, and separately control access to the page from a processor and from a direct memory access (DMA) request based on one or more fields of the stored page table entry. Other embodiments are disclosed and claimed.

Abstract: Embodiments of apparatuses, methods, and systems for virtualization of interprocessor interrupts are disclosed. In an embodiment, an apparatus includes a plurality of processor cores; an interrupt controller register; and logic to, in response to a write from a virtual machine to the interrupt controller register, record an interprocessor interrupt in a first data structure configured by a virtual machine monitor and send a notification of the interprocessor interrupt to at least one of the plurality of processor cores.

Abstract: Graphics processing systems and methods are described. A graphics processing apparatus may comprise one or more graphics processing engines, a memory, a memory management unit (MMU) including a GPU second level page table and GPU dirty bit tracking, and a provisioning agent to receive a request from a virtual machine monitor (VMM) to provision a subcluster of graphics processing apparatuses, the subcluster including a plurality of graphics processing engines from a plurality of graphics processing apparatuses connected using a scale-up fabric, provision the scale-up fabric to route data within the subcluster of graphics processing apparatuses, and provision a plurality of resources on the graphics processing apparatus for the subcluster based on the request from the VMM.

Abstract: Methods and apparatus for an accelerator controller hub (ACH). The ACH may be a stand-alone component or integrated on-die or on package in an accelerator such as a GPU. The ACH may include a host device link (HDL) interface, one or more Peripheral Component Interconnect Express (PCIe) interfaces, one or more high performance accelerator link (HPAL) interfaces, and a router, operatively coupled to each of the HDL interface, the one or more PCIe interfaces, and the one or more HPAL interfaces. The HDL interface is configured to be coupled to a host CPU via an HDL link and the one or more HPAL interfaces are configured to be coupled to one or more HPALs that are used to access high performance accelerator fabrics (HPAFs) such as NVlink fabrics and CCIX (Cache Coherent Interconnect for Accelerators) fabrics. Platforms including ACHs or accelerators with integrated ACHs support RDMA transfers using RDMA semantics to enable transfers between accelerator memory on initiators and targets without CPU involvement.

Abstract: Techniques and mechanisms for an input-output memory management module (IOMMU) to indicate to software whether a page request by an endpoint device is to be serviced. In an embodiment, the IOMMU receives from the endpoint device a response to an invalidation wait message. Based on the response, the IOMMU provides first information which indicates to software that page requests have been flushed from the endpoint device. Page request message from the endpoint device are compatible with an interface standard which also comprises a stop marker message type. The first information is provided independent of the endpoint device providing any message which is of the stop marker message type. In another embodiment, the first information includes a drain marker generated by the IOMMU, or a snapshot of an address corresponding to an end of a page request queue.

Abstract: Techniques for scalable virtualization of an Input/Output (I/O) device are described. An electronic device composes a virtual device comprising one or more assignable interface (AI) instances of a plurality of AI instances of a hosting function exposed by the I/O device. The electronic device emulates device resources of the I/O device via the virtual device. The electronic device intercepts a request from the guest pertaining to the virtual device, and determines whether the request from the guest is a fast-path operation to be passed directly to one of the one or more AI instances of the I/O device or a slow-path operation that is to be at least partially serviced via software executed by the electronic device. For a slow-path operation, the electronic device services the request at least partially via the software executed by the electronic device.

Abstract: In one embodiment, a processor comprises: a first configuration register to store a pointer to a process address space identifier (PASID) table; and an execution circuit coupled to the first configuration register. The execution circuit, in response to a first instruction, is to obtain command data from a first location identified in a source operand of the first instruction, obtain a PASID table handle from the command data, access a first entry of the PASID table using the pointer from the first configuration register and the PASID table handle to obtain a PASID value, insert the PASID value into the command data, and send the command data to a device coupled to the processor. Other embodiments are described and claimed.

Abstract: A method and system of host to guest (H2G) notification are disclosed. H2G is provided via an instruction. The instruction is a send user inter-processor interrupt instruction. An exemplary processor includes decoder circuitry to decode a single instruction and execute the decoded single instruction according to the at least the opcode to cause a host to guest notification from a virtual device running in a host machine on the first physical processor to a virtual device driver running on a virtual processor in a guest machine on a second physical processor.

Abstract: Techniques for region-aware memory bandwidth allocation control are described. In an embodiment, an apparatus includes a processing core and control circuitry. The processing core is to execute a plurality of threads. The control circuitry is to control use of memory bandwidth per memory region and per thread.

Abstract: An embodiment of an apparatus comprises decode circuitry to decode a single instruction, the single instruction to include a field for an identifier of a first source operand, a field for an identifier of a destination operand, and a field for an opcode, the opcode to indicate execution circuitry is to program a user timer, and execution circuitry to execute the decoded instruction according to the opcode to retrieve timer program information from a location indicated by the first source operand, and program a user timer indicated by the destination operand based on the retrieved timer program information. Other embodiments are disclosed and claimed.

Abstract: Methods and apparatus for PASID-based routing extension for Scalable IOV systems. The system may include a Central Processing Unit (CPU) operatively coupled to a scalable Input/Output Virtualization (IOV) device via an in-line device such as a smart controller or accelerator. A Control Process Address Space Identifier (C-PASID) associated with a first memory space is implemented in an Assignable Device Interface (ADI) for the IOV device. The ADI also implements a Data PASID (D-PASID) associated with a second memory space in which data are stored. The C-PASID is used to fetch a descriptor in the first memory space and the D-PASID is employed to fetch data in the second memory space. A hub embedded on the in-line device or implemented as a discrete device is used to steer memory access requests and/or fetches to the CPU or to the in-line device using the C-PASID and D-PASID.

Abstract: In one embodiment, a data mover accelerator is to receive, from a first agent having a first address space and a first process address space identifier (PASID) to identify the first address space, a first job descriptor comprising a second PASID selector to specify a second PASID to identify a second address space. In response to the first job descriptor, the data mover accelerator is to securely access the first address space and the second address space. Other embodiments are described and claimed.

Abstract: An embodiment of an integrated circuit may comprise an array of hardware counters, and circuitry communicatively coupled to the array of hardware counters, the circuitry to count accesses to one or more selected pages of a memory with the array of hardware counters. Other embodiments are disclosed and claimed.

Abstract: Embodiments of exitless guest to host (G2H) notification are described. In some embodiments, G2H is provided via an instruction. An exemplary processor includes decoder circuitry to decode a single instruction, the single instruction to include a field for an opcode; and execution processing resources to execute the decoded single instruction according to the at least the opcode to cause an exitless guest to host notification from a virtual processor to a physical or virtual processor.

Abstract: An embodiment of an integrated circuit may comprise an array of hardware counters, and circuitry communicatively coupled to the array of hardware counters, the circuitry to count accesses to one or more selected pages of a memory with the array of hardware counters. Other embodiments are disclosed and claimed.

Abstract: Embodiments are directed to providing a secure address translation service. An embodiment of a system includes memory for storage of data, an IOMMU coupled to the memory, and a host-to-device link to couple the IOMMU with one or more devices and to operate as a translation agent on behalf of one or more devices in connection with memory operations relating to the memory, including receiving a translated request from a discrete device via the host-to-device link specifying a memory operation and a physical address within the memory pertaining to the memory operation, determining page access permissions assigned to a context of the discrete device for a physical page of the memory within which the physical address resides, allowing the memory operation to proceed when the page access permissions permit the memory operation, and blocking the memory operation when the page access permissions do not permit the memory operation.

Abstract: Graphics processing systems and methods are described. A graphics processing apparatus may comprise one or more graphics processing engines, a memory, a memory management unit (MMU) including a GPU second level page table and GPU dirty bit tracking, and a provisioning agent to receive a request from a virtual machine monitor (VMM) to provision a subcluster of graphics processing apparatuses, the subcluster including a plurality of graphics processing engines from a plurality of graphics processing apparatuses connected using a scale-up fabric, provision the scale-up fabric to route data within the subcluster of graphics processing apparatuses, and provision a plurality of resources on the graphics processing apparatus for the subcluster based on the request from the VMM.

Abstract: In one embodiment, a processor comprises: a first configuration register to store quality of service (QoS) information for a process address space identifier (PASID) value associated with a first process; and an execution circuit coupled to the first configuration register, where the execution circuit, in response to a first instruction, is to obtain command data from a first location identified in a source operand of the first instruction, insert the QoS information and the PASID value into the command data, and send a request comprising the command data to a device coupled to the processor, to enable the device to use the QoS information of a plurality of requests to manage sharing between a plurality of processes. Other embodiments are described and claimed.

Abstract: Disclosed embodiments relate to systems and methods to skip inconsequential matrix operations. In one example, a processor includes decode circuitry to decode an instruction having fields to specify an opcode and locations of first source, second source, and destination matrices, the opcode indicating that the processor is to multiply each element at row M and column K of the first source matrix with a corresponding element at row K and column N of the second source matrix, and accumulate a resulting product with previous contents of a corresponding element at row M and column N of the destination matrix, the processor to skip multiplications that, based on detected values of corresponding multiplicands, would generate inconsequential results; scheduling circuitry to schedule execution of the instruction; and execution circuitry to execute the instructions as per the opcode.

Abstract: Graphics processing systems and methods are described. A graphics processing apparatus may comprise one or more graphics processing engines, a memory, a memory management unit (MMU) including a GPU second level page table and GPU dirty bit tracking, and a provisioning agent to receive a request from a virtual machine monitor (VMM) to provision a subcluster of graphics processing apparatuses, the subcluster including a plurality of graphics processing engines from a plurality of graphics processing apparatuses connected using a scale-up fabric, provision the scale-up fabric to route data within the subcluster of graphics processing apparatuses, and provision a plurality of resources on the graphics processing apparatus for the subcluster based on the request from the VMM.