摘要:
A technique for transferring data in a digital signal processing system is described. In one example, the digital signal processing system comprises a number of hardware peripherals, each connected to a memory access controller and each configured to read data from a memory device, perform one or more operations on the data, and write data to the memory device. To avoid hardwiring the hardware peripherals together, and to provide a configurable digital signal processing system, a multi-threaded processor controls the transfer of data between the hardware peripherals and the memory. Each processor thread is allocated to a memory access channel, and the threads are configured to detect an occurrence of an event and, responsive to this, control the memory access controller to enable a selected hardware peripheral to read data from or write data to the memory device via its memory access channel.
摘要:
A data processing system comprising a processing unit, a first memory, and a second memory, wherein the data processing system is arranged to hardware protect the second memory when a write access to the first memory is executed, wherein the processing unit is arranged to execute a program having at least one jump instruction and at least one return instruction, wherein the processing unit is arranged to store a program stack in the first memory, wherein the processing unit is arranged to store a return address on the program stack and to store a return address copy in the second memory when the at least one jump instruction is executed, and wherein the processing unit is arranged to compare the return address with the return address copy when the at least one return instruction is executed.
摘要:
A method of performing data transactions in a high performance persistent memory comprising, with a processor, updating data by writing new data to non-volatile memory (NVM) and receiving a done signal from a transaction accelerator communicatively coupled to the NVM. An apparatus for high performance persistent memory, comprising a processor, a memory controller communicatively coupled to the processor, and non-volatile memory communicatively coupled to the memory controller and processor, the non-volatile memory comprising an ACID transaction accelerator, in which the processor updates data on the non-volatile memory (NVM) by writing new data to the NVM, and receives a done signal from the an ACID transaction accelerator when the data has been updated.
摘要:
A technique for transferring data in a digital signal processing system is described. In one example, the digital signal processing system comprises a number of hardware peripherals, each connected to a memory access controller and each configured to read data from a memory device, perform one or more operations on the data, and write data to the memory device. To avoid hardwiring the hardware peripherals together, and to provide a configurable digital signal processing system, a multi-threaded processor controls the transfer of data between the hardware peripherals and the memory. Each processor thread is allocated to a memory access channel, and the threads are configured to detect an occurrence of an event and, responsive to this, control the memory access controller to enable a selected hardware peripheral to read data from or write data to the memory device via its memory access channel.
摘要:
A technique for executing a task sequence on a computing system comprising a multiple task processor having an on-chip memory and further comprising an external memory connected to the multiple task processor is provided. A method implementation of the technique comprises transferring load module data from the external memory into the on-chip memory in order to generate a load module sequence within the on-chip memory, wherein the generation of a load module of the load module sequence comprises the following processes: determining which parts of the load module are currently stored within the on-chip memory, and transferring only load module data from the external memory into the on-chip memory for parts of the load module which are currently not stored within the on-chip memory, wherein each load module of the load module sequence is generated within an individual address range of the on-chip memory which is chosen in dependence on the load module sequence. The method implementation further comprises executing the task sequence by running the load module sequence.
摘要:
A conditional processing method and apparatus for efficient memory management are provided. A conditional processing method includes generating a parse tree by loading a plurality of nodes of data structured based on a declarative description language in a memory in series; evaluating, when a parsing switch node having an attribute describing a condition for conditional processing exists among the nodes, child nodes of the parsing switch node according to the attribute; loading only the child nodes fulfilling the attribute in the memory; and outputting the child nodes retained on the memory.
摘要:
A conditional processing method and apparatus for efficient memory management are provided. A conditional processing method of the present inventions includes generating a parse tree by loading a plurality of nodes of data structured based on a declarative description language in a memory in series; evaluating, when a parsing switch node having an attribute describing a condition for conditional processing exists among the nodes, child nodes of the parsing switch node according to the attribute; loading only the child nodes fulfilling the attribute in the memory; and outputting the child nodes retained on the memory.
摘要:
The disclosed heterogeneous processor compresses information to more efficiently store the information in a system memory coupled to the processor. The heterogeneous processor includes a general purpose processor core coupled to one or more processor cores that exhibit an architecture different from the architecture of the general purpose processor core. In one embodiment, the processor dedicates a processor core other than the general purpose processor core to memory compression and decompression tasks. In another embodiment, system memory stores both compressed information and uncompressed information.
摘要:
A parallel processor (21) capable of exhibiting a high processing performance, which when receiving as input access requests generating page faults to sub-banks (27 1 , 27 2 , 27 3 ) from a plurality of processor elements (23 1 , 23 2 , ...23 n ) connected to a common bus (22) and another access request is input while data is being transferred between sub-banks (27 1 , 27 2 , 27 3 ) and an external memory (7) via an external access bus (26) in response to the input access requests, a shared memory (24) stores the other access request in a request queue and makes a control circuit execute the stored access request when the stored access request does not generate a page fault.