摘要:
A multi-processor system 8 includes multiple processing devices, including DSPs (10), processor units (MPUs) (21), co-processors (30) and DMA channels (31). Some of the devices may include internal MMUs (19, 32) which allows the device (10, 21, 30, 31) to work with a large virtual address space mapped to an external shared memory (20). The MMUs (19, 32) may perform the translation between a virtual address and the physical address associated with the external shared memory (20). Access to the shared memory (20) is controlled using a unified memory management system.
摘要:
A processor (e.g., a co-processor) comprising a decoder adapted to decode instructions from a first instruction set in a first mode and a second instruction set in a second mode. A pre-decoder coupled to the decoder, and operates in parallel with the decoder, determines the mode of operation for the decode logic for subsequent instructions. In particular, the decode logic operating in a current mode concurrently with the pre-decoder detecting a predetermined prefix, which indicates a subsequent instruction is a system command. Upon detecting this predetermined prefix, the decoder decodes the system command accordingly.
摘要:
A method is disclosed that comprises determining whether a data subsystem is to operate as cache memory or as scratchpad memory in which line fetches from external memory are suppressed and programming a control bit to cause the data subsystem to be operated as either a cache or scratchpad memory depending on the determination. Other embodiments are disclosed herein as well.
摘要:
A digital system has at least one processor, with an associated multi-segment cache memory circuit. A single global validity circuit (VIG) is connected to the memory circuit and is operable to indicate if any segment of the multiple segments holds valid data. Block circuitry is operable to transfer data from a pre-selected region of the secondary memory to a particular segment of the plurality of segments and to assert the global valid bit at the completion of a block transfer. Direct memory access (DMA) circuitry is connected to the memory cache for transferring data between the memory cache and a selectable region of a secondary memory and is also operable to assert the global valid bit at the completion of a DMA block transfer.
摘要:
A wireless data platform comprises a plurality of processors. Channels of communication are set up between processors such that they may communicate information as tasks are performed. A dynamic cross compiler executed on one processor compiles code into native processing code for another processor. A dynamic cross linker links the compiled code for other processor. Native code may also be downloaded to the platform through use of a JAVA Bean (or other language type) which encapsulates the native code. The JAVA Bean can be encrypted and digitally signed for security purposes.
摘要:
A processor comprising fetch logic adapted to fetch a set of instructions from memory, the set comprising a subset of instructions. The processor further comprises decode logic coupled to the fetch logic and adapted to process the set of instructions, and a clock coupled to the decode logic. When processed, an instruction from the set causes the clock to increment a counter external to the processor while the subset is processed. A status of the counter is manipulated to determine an efficiency level pertaining to the subset of instructions.
摘要:
A system comprising a counter adapted to monitor the memory consumption of the allocated memory resources. Upon reaching or surpassing the memory resource threshold provided, the counter may indicate the need for garbage collection. The garbage collector assesses the memory and releases memory resources that are consumed by the programs but are not needed anymore. The recycled memory resources are thus provided to the programs and the counter is updated accordingly. In addition, the system may also include instructions requesting memory resources. After detecting such instructions, the memory usage counter is updated either by the exact amount of memory allocated or the estimated amount of memory allocated. The counter may be implemented in hardware or in software.
摘要:
A transport packet parser (42) includes a transport packet header decoder (50) for identifying a packet identifier (PID) and continuity counter (CC) associated with a current packet. The PID along with an enable (En) bit is input to an PID associative memory (52) in search mode to identify an address associated with the PID. The address is used to access a CC associated with a previous packet for the same PID in a random access memory (62). The previous continuity counter is used along with other header information to determine whether the current packet satisfies predetermined criteria. If so, the packet is passed to a transport packet buffer for further processing.
摘要:
A digital system and method of operation is provided in which several processors (400[]) are connected to a shared resource (432). Each processor has a translation lookaside buffer (TLB) (310[]) that contains recently used page entries that each includes an access priority value. Access priority values are assigned to regions of address space, typically pages, according to the program or data that is stored on a given page. Access priority values are maintained in page tables with address translations, such that when a translated page address is loaded into a TLB, the access priority associated with that page is included in the TLB page entry. Arbitration circuitry (430) is connected to receive a request signal from each processor along with an access priority value (353[]) from each TLB in response to the requested address. The arbitration circuitry is operable to schedule access to the shared resource according to the access priority values provided by the TLBs.
摘要:
A multiprocessor system (10) includes a plurality of processing modules, such as MPUs (12), DSPs (14), and coprocessors/DMA channels (16). Power management software (38) in conjunction with profiles (36) for the various processing modules and the tasks to executed are used to build scenarios which meet predetermined power objectives, such as providing maximum operation within package thermal constraints or using minimum energy. Actual activities associated with the tasks are monitored during operation to ensure compatibility with the objectives. The allocation of tasks may be changed dynamically to accommodate changes in environmental conditions and changes in the task list. Temperatures may be computed at various points in the multiprocessor system by monitoring activity information associated with various subsystems. The activity measurements may be used to compute a current power dissipation distribution over the die. If necessary, the tasks in a scenario may be adjusted to reduce power dissipation. Further, activity counters may be selectively enabled for specific tasks in order to obtain more accurate profile information.