Abstract: This disclosure describes various examples of a system which uses a multi-bank, multi-port shared memory system that may be implemented as part of a system on a chip. The shared memory system may have particular applicability in the context of an artificial reality system, and may be designed to have distributed or varied latency for one or more memory banks and/or one or more components or subsystems within the system on a chip. The described shared memory system may be logically a single entity, but physically may have multiple memory banks, each accessible by any of a number of components or subsystems. In some examples, the memory system may enable concurrent, common, and/or shared access to memory without requiring, in some situations, full locking or arbitration.

Abstract: This disclosure describes various examples of a system which uses a multi-bank, multi-port shared memory system that may be implemented as part of a system on a chip. The shared memory system may have particular applicability in the context of an artificial reality system, and may be designed to have distributed or varied latency for one or more memory banks and/or one or more components or subsystems within the system on a chip. The described shared memory system may be logically a single entity, but physically may have multiple memory banks, each accessible by any of a number of components or subsystems. In some examples, the memory system may enable concurrent, common, and/or shared access to memory without requiring, in some situations, full locking or arbitration.

Abstract: This disclosure describes various examples of a system which uses a multi-bank, multi-port shared memory system that may be implemented as part of a system on a chip. The shared memory system may have particular applicability in the context of an artificial reality system, and may be designed to have distributed or varied latency for one or more memory banks and/or one or more components or subsystems within the system on a chip. The described shared memory system may be logically a single entity, but physically may have multiple memory banks, each accessible by any of a number of components or subsystems. In some examples, the memory system may enable concurrent, common, and/or shared access to memory without requiring, in some situations, full locking or arbitration.

Abstract: This disclosure describes various examples of a system which uses a multi-bank, multi-port shared memory system that may be implemented as part of a system on a chip. The shared memory system may have particular applicability in the context of an artificial reality system, and may be designed to have distributed or varied latency for one or more memory banks and/or one or more components or subsystems within the system on a chip. The described shared memory system may be logically a single entity, but physically may have multiple memory banks, each accessible by any of a number of components or subsystems. In some examples, the memory system may enable concurrent, common, and/or shared access to memory without requiring, in some situations, full locking or arbitration.

Abstract: The disclosure describes techniques for hierarchical power management of memory of an artificial reality system to reduce power consumption by the memory. An example device may be a peripheral device configured to generate artificial reality content for display or a head-mounted display unit (HMD) configured to output artificial reality content for display. The device includes memory divided into multiple memory blocks configurable to operate in a plurality of power modes. The device also includes memory block controllers controlling memory blocks. Each memory block controller controls which power mode in which the corresponding memory block is to operate, independent of any of the other memory blocks. The device includes a memory power controller configured to configure control registers of the memory block controllers to direct the memory block controllers to select one of the plurality of power modes for the memory blocks when the memory blocks are not being accessed.

Abstract: This disclosure describes various examples of a system which uses a multi-bank, multi-port shared memory system that may be implemented as part of a system on a chip. The shared memory system may have particular applicability in the context of an artificial reality system, and may be designed to have distributed or varied latency for one or more memory banks and/or one or more components or subsystems within the system on a chip. The described shared memory system may be logically a single entity, but physically may have multiple memory banks, each accessible by any of a number of components or subsystems. In some examples, the memory system may enable concurrent, common, and/or shared access to memory without requiring, in some situations, full locking or arbitration.

Abstract: A memory management system for managing a memory and includes a multi-stage memory management unit including control circuitry and cache memory. The cache memory may have a respective translation look-aside buffer for each stage of the multi-stage memory management unit. The control circuitry may be configured to generate a blank data request including a virtual address and information that specifies that data is not to be read from the memory, perform address translations based on the generated blank data request in multiple stages until a physical address is obtained, and discard the blank data request.

Abstract: A memory management unit may send page table walk requests to a page table descriptor in a main memory system and receive address translation information, with the page table walk requests including information that specifies an amount of further address translation information, and receive the further address translation information. The cache unit may intercept the page table walk requests, and modify content of the intercepted page table walk requests so the information that specifies the amount of further address translation information is extended from a first amount to a second amount greater than the first amount. The cache unit may store the second amount of further address translation information for use with data requests that are subsequent to a current data request, and provide the address translation information based upon an intercepted page table walk request being associated with address translation information already stored in the cache unit.

Abstract: The method of transferring data between a first and a second set of elements via a switch that includes a set of paths each associated with a weighting coefficient representing a data stream for each path. The method includes a credit flow control between the first set of elements and the switch and a credit flow control between the switch and the second set of elements. An available credit coefficient is computed for each element of the first set on the basis of a credit allocated by each element of the second set and of the weighting coefficient of each path.

Abstract: A memory management system for managing a memory and includes a multi-stage memory management unit including control circuitry and cache memory. The cache memory may have a respective translation look-aside buffer for each stage of the multi-stage memory management unit. The control circuitry may be configured to generate a blank data request including a virtual address and information that specifies that data is not to be read from the memory, perform address translations based on the generated blank data request in multiple stages until a physical address is obtained, and discard the blank data request.

Abstract: A memory management unit may send page table walk requests to a page table descriptor in a main memory system and receive address translation information, with the page table walk requests including information that specifies an amount of further address translation information, and receive the further address translation information. The cache unit may intercept the page table walk requests, and modify content of the intercepted page table walk requests so the information that specifies the amount of further address translation information is extended from a first amount to a second amount greater than the first amount. The cache unit may store the second amount of further address translation information for use with data requests that are subsequent to a current data request, and provide the address translation information based upon an intercepted page table walk request being associated with address translation information already stored in the cache unit.

Abstract: The method of transferring data between a first and a second set of elements via a switch that includes a set of paths each associated with a weighting coefficient representing a data stream for each path. The method includes a credit flow control between the first set of elements and the switch and a credit flow control between the switch and the second set of elements. An available credit coefficient is computed for each element of the first set on the basis of a credit allocated by each element of the second set and of the weighting coefficient of each path.