公开(公告)号：US11573870B2

公开(公告)日：2023-02-07

申请号：US16211924

申请日：2018-12-06

Applicant: Intel Corporation

Inventor： Manasi Deval ,  Nrupal Jani ,  Anjali Singhai Jain ,  Parthasarathy Sarangam ,  Mitu Aggarwal ,  Neerav Parikh ,  Kiran Patil ,  Rajesh M. Sankaran ,  Sanjay K. Kumar ,  Utkarsh Y. Kakaiya ,  Philip Lantz ,  Kun Tian

IPC: G06F11/20 ,  G06F9/455 ,  G06F13/42 ,  G06F9/48 ,  G06F13/40 ,  G06F15/173 ,  G06F3/06 ,  G06F13/16 ,  G06F9/50 ,  G06F21/60 ,  G06F9/54

Abstract: Examples may include a computing platform having a host driver to get a packet descriptor of a received packet stored in a receive queue and to modify the packet descriptor from a first format to a second format. The computing platform also includes a guest virtual machine including a guest driver coupled to the host driver, the guest driver to receive the modified packet descriptor and to read a packet buffer stored in the receive queue using the modified packet descriptor, the packet buffer corresponding to the packet descriptor.

公开(公告)号：US20220413909A1

公开(公告)日：2022-12-29

申请号：US17359409

申请日：2021-06-25

Applicant: Intel Corporation

Inventor： Utkarsh Y. Kakaiya ,  Rajesh M. Sankaran

IPC: G06F9/48

Abstract: Examples include techniques to enable quality of service (QoS) control for an accelerator device. Circuitry at an accelerator device implements QoS control responsive to receipt of a submission descriptor for a work request to execute a workload for an application hosted by a compute device coupled with the accelerator device. An example QoS control includes accepting the submission descriptor to a work queue at the accelerator device based on a work size of submission descriptor submissions of the application to the work queue over a unit of time not exceeding a submission rate threshold. The work queue is associated with an operational unit at the accelerator device to execute the workload based on information included in the submission descriptor. The work queue to be shared with at least one other application hosted by the compute device.

公开(公告)号：US11513979B2

公开(公告)日：2022-11-29

申请号：US17187271

申请日：2021-02-26

Applicant: Intel Corporation

Inventor： Rajesh M. Sankaran ,  David J. Harriman ,  Sean O. Stalley ,  Rupin H. Vakharwala ,  Ishwar Agarwal ,  Pratik M. Marolia ,  Stephen R. Van Doren

IPC: G06F13/16 ,  G06F13/42

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.

公开(公告)号：US11416281B2

公开(公告)日：2022-08-16

申请号：US16474978

申请日：2016-12-31

Applicant: Intel Corporation

Inventor： Rajesh M. Sankaran ,  Gilbert Neiger ,  Narayan Ranganathan ,  Stephen R. Van Doren ,  Joseph Nuzman ,  Niall D. McDonnell ,  Michael A. O'Hanlon ,  Lokpraveen B. Mosur ,  Tracy Garrett Drysdale ,  Eriko Nurvitadhi ,  Asit K. Mishra ,  Ganesh Venkatesh ,  Deborah T. Marr ,  Nicholas P. Carter ,  Jonathan D. Pearce ,  Edward T. Grochowski ,  Richard J. Greco ,  Robert Valentine ,  Jesus Corbal ,  Thomas D. Fletcher ,  Dennis R. Bradford ,  Dwight P. Manley ,  Mark J. Charney ,  Jeffrey J. Cook ,  Paul Caprioli ,  Koichi Yamada ,  Kent D. Glossop ,  David B. Sheffield

IPC: G06F9/48 ,  G06F9/30 ,  G06F9/38

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.

公开(公告)号：US11176059B2

公开(公告)日：2021-11-16

申请号：US16831976

申请日：2020-03-27

Applicant: Intel Corporation

Inventor： David M. Durham ,  Siddhartha Chhabra ,  Amy L. Santoni ,  Gilbert Neiger ,  Barry E. Huntley ,  Hormuzd M. Khosravi ,  Baiju V. Patel ,  Ravi L. Sahita ,  Gideon Gerzon ,  Ido Ouziel ,  Ioannis T. Schoinas ,  Rajesh M. Sankaran

IPC: G06F12/14 ,  G06F21/53 ,  G06F21/78 ,  G06F21/60 ,  G06F21/82 ,  G06F3/06

Abstract: In one embodiment, an apparatus comprises a processor to read a data line from memory in response to a read request from a VM. The data line comprises encrypted memory data. The apparatus also comprises a memory encryption circuit in the processor. The memory encryption circuit is to use an address of the read request to select an entry from a P2K table; obtain a key identifier from the selected entry of the P2K table; use the key identifier to select a key for the read request; and use the selected key to decrypt the encrypted memory data into decrypted memory data. The processor is further to make the decrypted memory data available to the VM. The P2K table comprises multiple entries, each comprising (a) a key identifier for a page of memory and (b) an encrypted address for that page of memory. Other embodiments are described and claimed.

公开(公告)号：US20210271481A1

公开(公告)日：2021-09-02

申请号：US17253053

申请日：2018-12-21

Applicant: Intel Corporation

Inventor： Kun Tian ,  Sanjay Kumar ,  Ashok Raj ,  Yi Liu ,  Rajesh M. Sankaran ,  Philip R. Lantz

IPC: G06F9/34 ,  G06F9/455 ,  G06F9/38 ,  G06F13/42 ,  G06F12/1009

Abstract: Process address space identifier virtualization uses hardware paging hint. The processing device (100) comprising: a processing core (110); and a translation circuit coupled to the processing core, the translation circuit to: receive a workload instruction from a guest application being executed by the processing device, the workload instruction comprising an untranslated guest process address space identifier (gPASID), a workload for an input/output (I/O) target device, and an identifier of a submission register on the I/O target device (410), access a paging data structure (PDS) associated with the guest application to retrieve a page table entry corresponding to the gPASID and the identifier of the submission register (420), determine a value of an I/O hint bit of the page table entry corresponding to the gPASID and the identifier of the submission register (430), responsive to determining that the I/O hint bit is enabled, keep the untranslated gPASID in the workload instruction (440), and provide the workload instruction to a work queue of the I/O target device (450)

公开(公告)号：US11106613B2

公开(公告)日：2021-08-31

申请号：US15940128

申请日：2018-03-29

Applicant: Intel Corporation

Inventor： Philip R. Lantz ,  Sanjay Kumar ,  Rajesh M. Sankaran ,  Saurabh Gayen

IPC: G06F13/364 ,  G06F13/24 ,  G06F9/50

Abstract: Embodiments of apparatuses, methods, and systems for highly scalable accelerators are described. In an embodiment, an apparatus includes an interface to receive a plurality of work requests from a plurality of clients and a plurality of engines to perform the plurality of work requests. The work requests are to be dispatched to the plurality of engines from a plurality of work queues. The work queues are to store a work descriptor per work request. Each work descriptor is to include all information needed to perform a corresponding work request.

公开(公告)号：US20210209037A1

公开(公告)日：2021-07-08

申请号：US17187271

申请日：2021-02-26

Applicant: Intel Corporation

Inventor： Rajesh M. Sankaran ,  David J. Harriman ,  Sean O. Stalley ,  Rupin H. Vakharwala ,  Ishwar Agarwal ,  Pratik M. Marolia ,  Stephen R. Van Doren

IPC: G06F13/16 ,  G06F13/42

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.

公开(公告)号：US11025544B2

公开(公告)日：2021-06-01

申请号：US16435328

申请日：2019-06-07

Applicant: Intel Corporation

Inventor： Pratik M. Marolia ,  Rajesh M. Sankaran ,  Ashok Raj ,  Nrupal Jani ,  Parthasarathy Sarangam ,  Robert O. Sharp

IPC: H04L12/747 ,  G06F13/28 ,  H04L12/861 ,  G06F12/1081 ,  H04L12/773

Abstract: A network interface controller can be programmed to direct write received data to a memory buffer via either a host-to-device fabric or an accelerator fabric. For packets received that are to be written to a memory buffer associated with an accelerator device, the network interface controller can determine an address translation of a destination memory address of the received packet and determine whether to use a secondary head. If a translated address is available and a secondary head is to be used, a direct memory access (DMA) engine is used to copy a portion of the received packet via the accelerator fabric to a destination memory buffer associated with the address translation. Accordingly, copying a portion of the received packet through the host-to-device fabric and to a destination memory can be avoided and utilization of the host-to-device fabric can be reduced for accelerator bound traffic.

公开(公告)号：US20210149815A1

公开(公告)日：2021-05-20

申请号：US17129496

申请日：2020-12-21

Applicant: Intel Corporation

Inventor： Saurabh Gayen ,  Philip R. Lantz ,  Dhananjay A. Joshi ,  Rupin H. Vakharwala ,  Rajesh M. Sankaran ,  Narayan Ranganathan ,  Sanjay Kumar

IPC: G06F12/10 ,  G06F12/0875 ,  G06F13/28 ,  G06F13/40 ,  G06F13/42

Abstract: Techniques for offload device address translation fetching are disclosed. In the illustrative embodiment, a processor of a compute device sends a translation fetch descriptor to an offload device before sending a corresponding work descriptor to the offload device. The offload device can request translations for virtual memory address and cache the corresponding physical addresses for later use. While the offload device is fetching virtual address translations, the compute device can perform other tasks before sending the corresponding work descriptor, including operations that modify the contents of the memory addresses whose translation are being cached. Even if the offload device does not cache the translations, the fetching can warm up the cache in a translation lookaside buffer. Such an approach can reduce the latency overhead that the offload device may otherwise incur in sending memory address translation requests that would be required to execute the work descriptor.