Abstract: Systems, methods, and computer-readable media are disclosed for virtualizing memory compute function resources to improve resource utilization and system performance are disclosed. A virtualized hypervisor may be provided that is configured to instantiate a respective memory function controller of each memory controller present in a system/device. The virtualized hypervisor may be further configured to maintain the memory function controllers and their corresponding memory compute functionality as shareable resources that can be allocated to system components upon request. The virtualized hypervisor may allocate a memory function controller and its corresponding memory compute functionality to a system component, and may further provide the system component with an exclusive grant of memory compute pages that can be utilized by the allocated memory function controller to execute a memory compute function to perform one or more operations (e.g., one or more computations) on behalf of the system component.

Abstract: Systems, methods, and computer-readable media are disclosed for virtualizing memory compute function resources to improve resource utilization and system performance are disclosed. A virtualized hypervisor may be provided that is configured to instantiate a respective memory function controller of each memory controller present in a system/device. The virtualized hypervisor may be further configured to maintain the memory function controllers and their corresponding memory compute functionality as shareable resources that can be allocated to system components upon request. The virtualized hypervisor may allocate a memory function controller and its corresponding memory compute functionality to a system component, and may further provide the system component with an exclusive grant of memory compute pages that can be utilized by the allocated memory function controller to execute a memory compute function to perform one or more operations (e.g., one or more computations) on behalf of the system component.

Abstract: A memory subsystem completes multiple read operations in parallel, utilizing the functionality of buffered memory modules in a daisy chain topology. A variable read latency is provided with each read command to enable memory modules to run independently in the memory subsystem. Busy periods of the memory device architecture are hidden by allowing data buses on multiple memory modules attached to the same data channel to run in parallel rather than in series and by issuing reads earlier than required to enable the memory devices to return from a busy state earlier. During scheduling of reads, the earliest received read whose target memory module is not busy is immediately issued at a next command cycle. The memory controller provides a delay parameter with each issued read. The number of cycles of delay is calculated to allow maximum utilization of the memory modules' data bus bandwidth without causing collisions on the memory channel.

Abstract: A method, computer program product and system for switching a defective signal line with a spare signal line without shutting down the computer system. A service processor monitors error correction code (ECC) check units configured to detect an error in a signal line. If an ECC check unit detects an error rate that exceeds a threshold, then the signal line with such an error rate may be said to be “defective.” The service processor configures switch control units in the driver/receiver pair associated with the defective signal line to be able to switch the defective signal line with a spare line upon receipt of a command from a memory controller switch control unit. In this manner, the system is not deactivated in order to switch a defective signal line with a spare line thereby reducing the time that the processor cannot send information to the memory buffers.

Abstract: A memory subsystem completes multiple read operations in parallel, utilizing the functionality of buffered memory modules in a daisy chain topology. A variable read latency is provided with each read command to enable memory modules to run independently in the memory subsystem. Busy periods of the memory device architecture are hidden by allowing data buses on multiple memory modules attached to the same data channel to run in parallel rather than in series and by issuing reads earlier than required to enable the memory devices to return from a busy state earlier. During scheduling of reads, the earliest received read whose target memory module is not busy is immediately issued at a next command cycle. The memory controller provides a delay parameter with each issued read. The number of cycles of delay is calculated to allow maximum utilization of the memory modules' data bus bandwidth without causing collisions on the memory channel.

Abstract: A system and method of recovering from errors in a data processing system. The data processing system includes one or more processor cores coupled to one or more memory controllers. The one or more memory controllers include at least a first memory interface coupled to a first memory and at least a second memory interface coupled to a second memory. In response to determining an error has been detected in the first memory, access to the first memory via the first memory interface is inhibited. Also, the first memory interface is locally restarted without restarting the second memory interface.

Abstract: A method and system for enabling directed temperature/power management at the DIMM-level and/or DRAM-level utilizing intelligent scheduling of memory access operations received at the memory controller. Hot spots within the memory subsystem, caused by operating the DIMMs/DRAMs above predetermined/preset threshold power/temperature values for operating a DIMM and/or a DRAM, are avoided/controlled by logic within the memory controller. The memory controller logic throttles the number/frequency at which commands (read/write operations) are issued to the specific DIMM/DRAM based on feedback data received from the specific DIMM/DRAM reaching the preset threshold power usage value.