Avago Technologies General IP (singapore) Pte. Ltd.

Lsi Corporation

Robert L Sheffield

Horia Cristian Simionescu

James A Rizzo

Gerald Edward Smith

Naveen Krishnamurthy

Adam Weiner

Rajeev Srinivasa Murthy

Mark Ish

Timothy E Hoglund

Gerald James Fredin

Andrew J Spry

Kapil Sundrani

Anant Baderdinni

Ross Zwisler

Kenneth J Gibson

James K Yu

Anand Narayanamurthy

A controller includes a virtual memory mapped to device-side Peripheral component interconnect express address space includes virtual buffers allocation for each data transfer. Each virtual buffer is associated with a scatter/gather list entry in a host memory. The controller executes direct transfers between Peripheral component interconnect express devices and host memory without introducing address mapping dependencies between the host and device domains.

Decoupling host and device address maps for a peripheral component interconnect express controller

Methods and structure for host-side device drivers for Redundant Array of Independent Disks (RAID) systems. One system includes a processor and memory of a host, which implement a device driver. The device driver receives an Input/Output (I/O) request from an Operating System (OS) of the host, translates Logical Block Addresses (LBAs) from the received request into physical addresses at multiple storage devices, generates child I/O requests directed to the physical addresses based on the received request, and accesses an address lock system at a RAID controller to determine whether the physical addresses are accessible. If the physical addresses are accessible, the device driver reserves the physical addresses by updating the address lock system, and directs the child I/O requests to a hardware path at the RAID controller for handling single-strip I/O requests. If the physical addresses are not accessible, the device driver delays processing of the child I/O requests.

Host-based device drivers for enhancing operations in redundant array of independent disks systems

Methods for managing region lock in a storage controller are disclosed. Upon receiving in a hardware based region lock management circuit a data request, the region lock management circuit determines a lock type of a region lock specified in the data request and conditionally creates a region lock data structure, wherein the region lock data structure is conditionally created based on the type of the region lock requested. The region lock management circuit further determines whether the data request is requesting access to at least a portion of a locked data region and the lock type of the locked data region. If the data request is requesting access to at least a portion of the locked data region and the lock type of the locked data region permits firmware diversion, the region lock management circuit diverts the data request to a storage controller firmware processor for further processing.

Hardware based cache scan with divert node handling

A region lock (RL) method and system for ensuring data integrity is disclosed. The method and system in accordance with the present disclosure works in conjunction with a balanced-tree based RL scheme. By eliminating steps and checks that in most cases are unnecessary, the relatively high overhead associated with the balanced-tree based RL scheme may be reduced. For instance, the solution in accordance with the present disclosure may utilize a hash table to determine whether RL overlap checks may be bypassed for certain I/O commands. Since the new solution requires very little processing, therefore by reducing unnecessary RL overlap checks, RL overhead may be dramatically reduced and may lead to significant increases in overall system performance.

Hash-based region locking

Methods and structure within a storage controller for using region locks to efficiently divert an I/O request received from an attached host system to one of multiple processing stacks in the controller. A region lock module within the controller allows each processing stack to request a region lock for a range of block addresses of the storage devices. A divert-type lock request may be established to identify a range of block addresses for which I/O requests should be diverted to a particular one of the multiple processing stacks.

Methods and structure for using region locks to divert I/O requests in a storage controller having multiple processing stacks

A storage controller system is provided for the monitoring of fast path processing of I/Os to a storage device where the storage controller system allows for processing of I/Os to be monitored through the use of counters in a storage controller based upon the type of I/O issued to the storage controller as well as the conditions associated with the I/O, while providing a bitmap and associated divert bits and counters to monitor the processing of the I/Os in the storage controller. Methods for monitoring the processing of I/Os issued to the storage controller are also provided where the processing of the I/Os is based upon the type fast path I/Os issued to the storage controller and the conditions associated with the issued I/Os while providing a bitmap and associated divert bits and counters to monitor the processing of the I/Os in the storage controller, are also disclosed.

Confirmed divert bitmap to synchronize raid firmware operations with fast-path hardware I/O processing

A first I/O transaction request is sent to a storage controller for processing by firmware running on the storage controller. A second I/O transaction request is sent to storage hardware without further processing by the firmware running on the storage controller. Since the firmware did not process the second I/O transaction request, information associated with the second I/O transaction is stored in in a circular buffer accessible to the firmware running on the storage controller. The firmware running on the storage controller reads, from the circular buffer, the information associated with the second I/O transaction that was stored in the circular buffer.

Sharing of bypassed I/O transaction information

Methods and apparatus for improved building of a hot spare storage device in a RAID storage system while avoiding reading of stale data from a failed storage device. In the recovery mode of the failed device, all data is write protected on the failed device. A RAID storage controller may copy as much readable data as possible from the failed device to the hot spare storage device. Unreadable data may be rebuilt using redundant information of the logical volume. Write requests directed to the failed device cause the addressed logical block address (LBA) to be marked as storing stale data. When a read request is directed to such a marked LBA, the read request returns an error status from the failed device to indicate that the data is stale. The RAID controller then rebuilds the now stale data for that LBA from redundant information of the logical volume.

Methods and apparatus for marking writes on a write-protected failed device to avoid reading stale data in a RAID storage system

The present invention is a method for providing multi-pathing via Small Computer System Interface Input/Output (SCSI I/O) referral between an initiator and a storage cluster which are communicatively coupled via a network. The method includes receiving an input/output (I/O) at a first target device from the initiator via the network. The I/O includes a data request. The method further includes transmitting a SCSI I/O referral list to the initiator when data included in the data request is not stored on the first target device, but is stored on a second target device. The referral list includes first and second port identifiers for identifying first and second ports of the second target device respectively. The first and second ports of the target device are identified as access ports for accessing the data requested in the data request.

Transport agnostic SCSI I/O referrals

Methods and structure are disclosed for improved processing of fast path I/O requests in a storage controller utilizing version information embedded in the fast path I/O requests. The version information allows the storage controller to determine if the mapping information utilized by the host system in generating a fast path I/O request specifies the mapping information utilized by the storage controller. The controller comprises a fast path I/O request processing circuit tightly coupled with host system drivers for fast processing of requests directed to storage devices of a logical volume. The controller also comprises a logical volume I/O processing software stack for processing I/O requests from a host system directed to a logical volume. If the mapping information utilized by the host does not match the mapping information utilized by the storage controller, fast path I/O requests are transferred to the I/O request processing stack for subsequent processing.

Growing community of inventors

Longmont, CO, United States of America

Robert L Sheffield