Abstract: A system for providing odd modulus memory channel interleaving may include a dynamic random access memory (DRAM) system and a system on chip (SoC). The SoC comprises a first memory controller, a second memory controller, and a symmetric memory channel interleaver. The first memory controller is electrically coupled to a first DRAM module via a first memory bus. The second memory controller is electrically coupled to a second DRAM module and a third DRAM module via a second memory bus. The symmetric memory channel interleaver is configured to uniformly distribute DRAM traffic to the first memory controller and the second memory controller. The first memory controller provides a first interleaved channel to the first DRAM module via the first memory bus. The second memory controller provides a second interleaved channel to the second DRAM module via upper address bits on the second memory bus.

Abstract: Systems and methods are disclosed for expanding memory for a system on chip (SoC). A memory card is loaded in an expandable memory socket electrically and is coupled to a system on chip (SoC) via an expansion bus. The memory card comprises a first volatile memory device. In response to detecting the memory card, an expanded virtual memory map is configured. The expanded virtual memory map comprises a first virtual memory space associated the first volatile memory device and a second virtual memory space associated with a second volatile memory device electrically coupled to the SoC via a memory bus. One or more peripheral images associated with the second virtual memory space are relocated to a first portion of the first virtual memory space. A second portion of the first virtual memory space is configured as a block device for performing swap operations associated with the second virtual memory space.

Abstract: A memory having a redundancy area is operated in a normal mode and an error is detected. A selecting selects between in-line repair process and off-line repair. In-line repair applies a short term error correction, which remaps a fail address to a remapped memory area of the memory. An in-system repair is applied, for a one-time programmed remapping of the fail address to a redundancy area of the memory. In-system repair utilizes idle time of the memory to maintain valid memory content.

Abstract: Systems, methods, and computer programs are disclosed for method for reducing memory subsystem power. In an exemplary method, a system resource manager provides memory performance requirements for a plurality of memory clients to a double data rate (DDR) subsystem. The DDR subsystem and the system resource manager reside on a system on chip (SoC) electrically coupled to a dynamic random access memory (DRAM). A cache hit rate is determined of each of the plurality of memory clients associated with a system cache residing on the DDR subsystem. The DDR subsystem adjusts access to the DRAM based on the memory performance requirements received from the system resource manager and the cache hit rates of the plurality of memory clients.

Abstract: Systems, methods, and computer programs are disclosed for method for reducing memory subsystem power. In an exemplary method, a system resource manager provides memory performance requirements for a plurality of memory clients to a double data rate (DDR) subsystem. The DDR subsystem and the system resource manager reside on a system on chip (SoC) electrically coupled to a dynamic random access memory (DRAM). A cache hit rate is determined of each of the plurality of memory clients associated with a system cache residing on the DDR subsystem. The DDR subsystem controls a DDR clock frequency based on the memory performance requirements received from the system resource manager and the cache hit rates of the plurality of memory clients.

Abstract: Systems, methods, and computer programs are disclosed for method for reducing memory subsystem power. In an exemplary method, a system resource manager provides memory performance requirements for a plurality of memory clients to a double data rate (DDR) subsystem. The DDR subsystem and the system resource manager reside on a system on chip (SoC) electrically coupled to a dynamic random access memory (DRAM). A cache hit rate is determined of each of the plurality of memory clients associated with a system cache residing on the DDR subsystem. The DDR subsystem controls a DDR clock frequency based on the memory performance requirements received from the system resource manager and the cache hit rates of the plurality of memory clients.

Abstract: Providing memory training of dynamic random access memory (DRAM) systems using port-to-port loopbacks, and related methods, systems, and apparatuses are disclosed. In one aspect, a first port within a DRAM system is coupled to a second port via a loopback connection. A signal is sent to the first port from a System-on-Chip (SoC), and passed to the second port through the loopback connection. The signal is then returned to the SoC, where it may be examined by a closed-loop engine of the SoC. A result corresponding to a hardware parameter may be recorded, and the process may be repeated until an optimal result for the hardware parameter is achieved at the closed-loop engine. By using a port-to-port loopback configuration, the DRAM system parameters regarding timing, power, and other parameters associated with the DRAM system may be trained more quickly and with lower boot memory usage.

Abstract: Systems, methods, and computer programs are disclosed for method for reducing memory subsystem power. In an exemplary method, a system resource manager provides memory performance requirements for a plurality of memory clients to a double data rate (DDR) subsystem. The DDR subsystem and the system resource manager reside on a system on chip (SoC) electrically coupled to a dynamic random access memory (DRAM). A cache hit rate is determined of each of the plurality of memory clients associated with a system cache residing on the DDR subsystem. The DDR subsystem adjusts access to the DRAM based on the memory performance requirements received from the system resource manager and the cache hit rates of the plurality of memory clients.

Abstract: Providing memory training of dynamic random access memory (DRAM) systems using port-to-port loopbacks, and related methods, systems, and apparatuses are disclosed. In one aspect, a first port within a DRAM system is coupled to a second port via a loopback connection. A training signal is sent to the first port from a System-on-Chip (SoC), and passed to the second port through the loopback connection. The training signal is then returned to the SoC, where it may be examined by a closed-loop training engine of the SoC. A training result corresponding to a hardware parameter may be recorded, and the process may be repeated until an optimal result for the hardware parameter is achieved at the closed-loop training engine. By using a port-to-port loopback configuration, the DRAM system parameters regarding timing, power, and other parameters associated with the DRAM system may be trained more quickly and with lower boot memory usage.

Abstract: Systems, methods, and computer programs are disclosed for allocating memory in a hybrid parallel/serial memory system. One method comprises configuring a memory address map for a multi-rank memory system with a dedicated serial access region in a first memory rank and a dedicated parallel access region in a second memory rank. A request is received for a virtual memory page. If the request comprises a performance hint, the virtual memory page is selectively assigned to a free physical page in the dedicated serial access in the first memory rank and the dedicated parallel access region in the second memory rank.

Abstract: Power saving techniques for memory systems are disclosed. In particular, exemplary aspects of the present disclosure contemplate taking advantage of patterns that may exist within memory elements and eliminating duplicative data transfers. Specifically, if data is repetitive, instead of sending the same data repeatedly, the data may be sent only a single time with instructions that cause the data to be replicated at a receiving end to restore the data to its original repeated state. By reducing the amount of data that is transferred between a host and a memory element, power consumption is reduced.

Abstract: Systems and methods for improved flash memory performance in a portable computing device are presented. In a method, a value N corresponding to an amount of prefetch data to be retrieved from the flash memory is determined. An access request for a flash memory is received at a cache controller in communication with a cache memory. A determination is made whether the access request for the flash memory corresponds to a portion of data stored in the cache memory. If the access request for the flash memory corresponds to the portion of data, the portion of data is returned in response to the access request. Otherwise, an N amount of prefetch data is retrieved from the flash memory and stored in the cache memory. The value N is incremented based on a cache hit percentage for the cache memory.

Abstract: Systems, methods, and computer programs are disclosed for providing error detection or correction with flash cell mapping. One embodiment is a method comprising generating raw page data for a physical page in a main array of a flash memory device. The raw page data comprises less than a capacity of the physical page generated using a non-power-of-two flash cell mapping. One or more parity bits are generated for the raw page data using an error detection or correction scheme. The method stores the raw page data and the one or more parity bits in the physical page in the main array.

Abstract: Systems, methods, and computer programs are disclosed for providing non-volatile system memory with volatile memory program caching. One such method comprises storing an executable program in a non-volatile random access memory. In response to an initial launch of the executable program, the executable program is loaded from the non-volatile random access memory into a volatile memory cache for execution. In response to an initial suspension of the executable program, cache pages corresponding to the executable program are flushed into the non-volatile random access memory.

Abstract: Systems and methods are disclosed for providing logical-to-physical address translation for a managed NAND flash storage device. One embodiment is a system comprising a system on chip (SoC) electrically coupled to a volatile memory device. A direct memory access (DMA) controller is electrically coupled to the SoC. The DMA controller receives a logical address to be translated to a physical address associated with a managed NAND flash storage device. A cache controller is configured to fetch from the volatile memory device a portion of a logical-to-physical (L2P) address table comprising a compressed version of a L2P mapping for the logical address. A compression block is configured to decompress the compressed version of the L2P mapping to determine the physical address corresponding to the logical address.

Abstract: Systems, methods, and computer programs are disclosed for resolving dynamic random access memory (DRAM) defects. One embodiment is a system comprising a dynamic random access memory (DRAM) system electrically coupled to a system on chip (SoC). The SoC comprises a cache and a cache controller. The cache controller is configured to store corrected data for a failed physical codeword address associated with the DRAM in the cache and provide further access to the failed physical codeword address from the cache instead of the DRAM system.

Abstract: Systems and methods are disclosed for providing memory channel interleaving with selective power/performance optimization. One such method comprises configuring an interleaved zone for relatively higher performance tasks, a linear address zone for relatively lower power tasks, and a mixed interleaved-linear zone for tasks with intermediate performance requirements. A boundary is defined among the different zones using a sliding threshold address. The zones may be dynamically adjusted, and/or new zones dynamically created, by changing the sliding address in real-time based on system goals and application performance preferences. A request for high performance memory is allocated to a zone with lower power that minimally supports the required performance, or may be allocated to a low power memory zone with lower than required performance if the system parameters indicate a need for aggressive power conservation. Pages may be migrated between zones in order to free a memory device for powering down.

Abstract: Components of a portable computing device produce power supply voltage requests indicating requested power levels. In response to the power supply voltage requests, power multiplexers associated with the components select and couple corresponding voltage rails associated with two or more fixed-voltage power supplies to the requesting components. Power supplies may be activated and deactivated on an as-requested basis.

Abstract: Systems and methods for improved flash memory performance in a portable computing device are presented. In a method, a value N corresponding to an amount of prefetch data to be retrieved from the flash memory is determined. An access request for a flash memory is received at a cache controller in communication with a cache memory. A determination is made whether the access request for the flash memory corresponds to a portion of data stored in the cache memory. If the access request for the flash memory corresponds to the portion of data, the portion of data is returned in response to the access request. Otherwise, an N amount of prefetch data is retrieved from the flash memory and stored in the cache memory. The value N is incremented based on a cache hit percentage for the cache memory.

Abstract: Systems, methods, and computer programs, embodied in or as a memory management module, are disclosed for thermally controlling memory to increase its performance. One exemplary embodiment includes a memory, one or more processors, and a thermoelectric cooling device. The one or more processors access the memory via a memory controller electrically coupled to the memory. The thermoelectric cooling device is configured to thermally control the memory in response to a predicted change in temperature of the memory.