Intel Corporation

Sk Hynix Nand Product Solutions Corp.

Geetani R Edirisooriya

Pankaj Kumar

John P Lee

Mark A Schmisseur

Michael T Klinglesmith

Samantha J Edirisooriya

Bradley A Burres

Sivakumar Radhakrishnan

Eric Dahlen

Sin Sim Tan

Thomas M Slaight

Brian M Leitner

Aniruddha P Joshi

David C Lee

Marc Tim Jones

Matthew R Wilcox

Kenneth C Haren

Saptarshi Mondal

Roger C Jeppsen

Daniel Nemiroff

Robert W Strong

Rajeev K Nalawadi

Robert P Adler

Joseph Murray

Amber D Huffman

Deep K Buch

Jeffrey L Rabe

Jasper Balraj

John Lee

Aniruddha Joshi

In accordance with embodiments disclosed herein, there are provided methods, systems, mechanisms, techniques, and apparatuses for presentation of direct accessed storage under a logical drive model; for implementing a distributed architecture for cooperative NVM Data protection; data mirroring for consistent SSD latency; for boosting a controller's performance and RAS with DIF support via concurrent RAID processing; for implementing arbitration and resource schemes of a doorbell mechanism, including doorbell arbitration for fairness and prevention of attack congestion; and for implementing multiple interrupt generation using a messaging unit and NTB in a controller through use of an interrupt coalescing scheme.

Presentation of direct accessed storage under a logical drive model

A processing system includes a processor and a VM-to-VM communication accelerator circuit comprising a first interface device to support direct memory access (DMA) data transfers by the first VM, a register to store a reference to a primary physical function (PF) associated with the first interface device, wherein the first primary PF is associated with an access control table (ACT) specifying an access permission for the first VM with respect to a second VM, and a direct memory access (DMA) descriptor processing circuit to process, using a working queue associated with the first primary PF, a DMA descriptor referencing a request for a DMA data transfer between the first VM and the second VM, and execute, using the first interface device, the DMA data transfer based on the access permission.

Hardware-based virtual machine communication supporting direct memory access data transfer

Hardware-based virtual machine communication

In an embodiment, a processor for invalidating cache entries comprises: at least one processing unit; a processor cache; and direct cache unit. The direct cache unit is to receive, from a first device, a direct read request for data in a first cache entry in the processor cache; determine whether the direct read request is an invalidating read request; in response to a determination that the direct read request is an invalidating read request: send the data in the first cache entry directly from the processor cache to the first device without accessing a main memory; and invalidate the first cache entry in the processor cache. Other embodiments are described and claimed.

Invalidating reads for cache utilization in processors

An inbound sideband interface is provided to receive a message over a first sideband link, and parity logic is provided to calculate a parity bit for the message. Further, an outbound sideband interface is provided to forward the message to another device over a second sideband link. The second sideband link includes a plurality of data wires and a parity bit wire. The message is forwarded over at least some of the data wires and the parity bit is sent to the other device over the parity bit wire to correspond with the message.

Sideband parity handling

A method, apparatus, and system are disclosed. In one embodiment the method detects a temperature event in a processor and then modifies the bus frequency of an I/O bus coupled to an I/O controller hub in response to the temperature event. In another embodiment, the apparatus includes a temperature detection unit that detects a temperature event in a processor and, additionally, a bus frequency modification unit that will modify the bus frequency of an I/O bus in response to the temperature event.

Method, system, and apparatus to decrease CPU temperature through I/O bus throttling

A computer system provides a program access to a first register during real mode operation by using an index register and a data register, wherein the index register and the data register are located in real mode memory space and the first register is located outside of real mode memory space.

Index/data register pair for indirect register access

In some embodiments, a method and apparatus for maintaining data integrity following parity error detection are described. In one embodiment, the method includes the blockage of bus transactions in response to detection of a parity error. Once bus transactions are suspended, a parity error handler routine is invoked to perform parity error recovery according to stored transaction information regarding an error bus transaction of the detected parity error. In one embodiment, the stored information includes a bus master that caused the error, as well as an address associated with the corrupt data for which the parity error was asserted. In one embodiment, data logging is performed to track the bus masters associated with error bus transactions to enable identification of problematic or old hardware devices. Other embodiments are described and claims.

Apparatus and method for maintaining data integrity following parity error detection

Method and apparatus for supporting multi-function PCI devices in PCI bridges. Respective pre-fetch buffers are allocated in response to respective initial data transfer requests issued by a multi-function PCI device. A programmable buffer fill watermark is set up for each pre-fetch buffer. While a portion of data corresponding to the data transfer requests fill the pre-fetch buffers, the fill level of each buffer is monitored to determine if it meets or exceeds its buffer fill watermark. In response to such a condition, the multi-function PCI device is connected to the PCI bridge and a virtual buffer is mapped to the pre-fetch buffer. The pre-fetch buffer is then emptied. During subsequent data transfers, each of the pre-fetch buffer becomes filled, the PCI device is connected, and the virtual buffer is mapped to the filled buffer. The process is continued until all data corresponding to the original data transfer request is received by the multi-function PCI device.

Method and apparatus for managing buffers in PCI bridges

Growing community of inventors

Tempe, AZ, United States of America

Geetani R Edirisooriya