Abstract: One aspect of the application can provide a system and method for replacing a failing node with a spare node in a non-uniform memory access (NUMA) system. During operation, in response to determining that a node-migration condition is met, the system can initialize a node controller of the spare node such that accesses to a memory local to the spare node are to be processed by the node controller, quiesce the failing node and the spare node to allow state information of processors on the failing node to be migrated to processors on the spare node, and subsequent to unquiescing the failing node and the spare node, migrate data from the failing node to the spare node while maintaining cache coherence in the NUMA system and while the NUMA system remains in operation, thereby facilitating continuous execution of processes previously executed on the failing node.

Abstract: The disclosed embodiments disclose techniques for performing physical domain error isolation and recovery in a multi-domain system, where the multi-domain system includes two or more processor chips and one or more switch chips that provide connectivity and cache-coherency support for the processor chips, and the processor chips are divided into two or more distinct domains. During operation, one of the switch chips determines a fault in the multi-domain system. The switch chip determines an originating domain that is associated with the fault, and then signals the fault and an identifier for the originating domain to its internal units, some of which perform clearing operations that clear out all traffic for the originating domain without affecting the other domains of the multi-domain system.

Abstract: The disclosed embodiments disclose techniques for performing physical domain error isolation and recovery in a multi-domain system, where the multi-domain system includes two or more processor chips and one or more switch chips that provide connectivity and cache-coherency support for the processor chips, and the processor chips are divided into two or more distinct domains. During operation, one of the switch chips determines a fault in the multi-domain system. The switch chip determines an originating domain that is associated with the fault, and then signals the fault and an identifier for the originating domain to its internal units, some of which perform clearing operations that clear out all traffic for the originating domain without affecting the other domains of the multi-domain system.

Abstract: Some embodiments of the present invention provide a system that processes a request for a cache line in a multiprocessor system that supports a directory-based cache-coherence scheme. During operation, the system receives the request for the cache line from a requesting node at a home node, wherein the home node maintains directory information for all or a subset of the address space which includes the cache line. Next, the system performs an action at the home node, which causes a valid copy of the cache line to be sent to the requesting node. The system then completes processing of the request at the home node without waiting for an acknowledgment indicating that the requesting node received the valid copy of the cache line.

Abstract: Some embodiments of the present invention provide a system that processes a request for a cache line in a multiprocessor system that supports a directory-based cache-coherence scheme. During operation, the system receives the request for the cache line from a requesting node at a home node, wherein the home node maintains directory information for all or a subset of the address space which includes the cache line. Next, the system performs an action at the home node, which causes a valid copy of the cache line to be sent to the requesting node. The system then completes processing of the request at the home node without waiting for an acknowledgment indicating that the requesting node received the valid copy of the cache line.

Abstract: A processor employing device-specific memory address translation. In one embodiment, a processor may include a device interface configured to receive a memory access request from an input/output (I/O) device, where the request specifies a virtual memory address and a first requestor identifier (ID) that identifies the I/O device. The processor may also include an I/O memory management unit coupled to the device interface and configured to determine whether a virtual-to-physical memory address translation corresponding to the virtual memory address is stored within an I/O memory translation buffer. The I/O memory management unit may be further configured to determine whether a second requestor ID stored within the I/O memory translation buffer and corresponding to the memory address translation matches the first requestor ID. If the first and second requestor IDs do not match, the I/O memory management unit may disallow the memory access request and to signal an error condition.

Abstract: In one of the many embodiments disclosed herein, a method for dispatching read and write requests to a memory is disclosed which includes queuing at least one write request in a write queue and queuing an incoming read request in a read queue. The method also includes comparing the read request with at least one write request in the write queue to detect a matching write request, and if there is a matching write request, storing a write queue index of the matching write request as a first entry in an ordering queue. The method further includes dispatching the at least one write request to the memory in response to the first ordering queue entry.

Abstract: A memory interface unit for controlling a memory includes a read queue, a write queue, and an ordering circuit. The read queue stores read requests for the memory, and the write queue stores write requests for the memory. The ordering circuit is coupled to both the read queue and the write queue, and allows read and write requests to be dispatched to memory in an out-of-order manner with respect to each other to maximize performance without jeopardizing data coherency.

Abstract: A multiprocessor system includes a plurality of central processing units (CPUs) connected to one another by a system bus. Each CPU includes a cache controller to communicate with its cache, and a primary memory controller to communicate with its primary memory. When there is a cache miss in a CPU, the cache controller routes an address request for primary memory directly to the primary memory via the CPU as a speculative request without access the system bus, and also issues the address request to the system bus to facilitate data coherency. The speculative request is queued in the primary memory controller, which in turn retrieves speculative data from a specified primary memory address. The CPU monitors the system bus for a subsequent transaction that requests the specified data in the primary memory. If the subsequent transaction requesting the specified data is a read transaction that corresponds to the speculative address request, the speculative request is validated and becomes non-speculative.