| Patent application number | Description | Published |
|---|
| 20080209182 | MULTI-MODE DATA PROCESSING DEVICE AND METHODS THEREOF - A data processing device and methods thereof are disclosed. The data processing device can operate in three different modes. In a first, N-bit mode, the data processing device performs memory accesses based on N-bit values and performs arithmetic operations using N-bit values. In a second, hybrid N-bit/M-bit mode, the data processing device performs memory accesses based on M-bit values, where M is less than N, and performs arithmetic operations using N-bit values. In a third, M-bit mode, the data processing device performs memory accesses based on M-bit values and performs arithmetic operations using M-bit values. The three modes provide for compatibility with a wide range of applications. Further operation in the M-bit mode can provide a power savings when implementing applications compatible with that mode. | 08-28-2008 |
| 20080222382 | PERFORMANCE MONITORING DEVICE AND METHOD THEREOF - A performance monitoring device and method are disclosed. The device monitors performance events of a processor. A counter is adjusted in response to the occurrence of a particular performance event. The counter can be associated with a particular instruction address range, or a data address range, so that the counter is adjusted only when the performance event occurs at the instruction address range or the data address range. Accordingly, the information stored in the counter can be analyzed to determine if a particular instruction address range or data address range results in a particular performance event. Multiple counters, each associated with a different performance event, instruction address range, or data address range, can be employed to allow for a detailed analysis of which portions of a program lead to particular performance events. | 09-11-2008 |
| 20080222389 | INTERPROCESSOR MESSAGE TRANSMISSION VIA COHERENCY-BASED INTERCONNECT - A method includes communicating a first message between processors of a multiprocessor system via a coherency interconnect, whereby the first message includes coherency information. The method further includes communicating a second message between processors of the multiprocessor system via the coherency interconnect, whereby the second message includes interprocessor message information. A system includes a coherency interconnect and a processor. The processor includes an interface configured to receive messages from the coherency interconnect, each message including one of coherency information or interprocessor message information. The processor further includes a coherency management module configured to process coherency information obtained from at least one of the messages and an interrupt controller configured to generate an interrupt based on interprocessor message information obtained from at least one of the messages. | 09-11-2008 |
| 20080282251 | THREAD DE-EMPHASIS INSTRUCTION FOR MULTITHREADED PROCESSOR - A technique for scheduling execution of threads at a processor is disclosed. The technique includes executing a thread de-emphasis instruction of a thread that de-emphasizes the thread until the number of pending memory transactions, such as cache misses, associated with the thread are at or below a threshold. While the thread is de-emphasized, other threads at the processor that have a higher priority can be executed or assigned system resources. Accordingly, the likelihood of a stall in the processor is reduced. | 11-13-2008 |
| 20080288724 | METHOD AND APPARATUS FOR CACHE TRANSACTIONS IN A DATA PROCESSING SYSTEM - A plurality of new snoop transaction types are described. Some include address information in the requests, and others include cache entry information in the requests. Some responses include tag address information, and some do not. Some provide tag address content on the data bus lines during the data portion of the transaction. These new snoop transaction types are very helpful during debug of a data processing system. | 11-20-2008 |
| 20080288725 | METHOD AND APPARATUS FOR CACHE TRANSACTIONS IN A DATA PROCESSING SYSTEM - A plurality of new snoop transaction types are described. Some include address information in the requests, and others include cache entry information in the requests. Some responses include tag address information, and some do not. Some provide tag address content on the data bus lines during the data portion of the transaction. These new snoop transaction types are very helpful during debug of a data processing system. | 11-20-2008 |
| 20090019232 | SPECIFICATION OF COHERENCE DOMAIN DURING ADDRESS TRANSLATION - A processing system includes a plurality of coherency domains and a plurality of coherency agents. Each coherency agent is associated with at least one of the plurality of coherency domains. At a select coherency agent of the plurality of coherency agents, an address translation for a coherency message is performed using a first memory address to generate a second memory address. A select coherency domain of the plurality of coherency domains associated with the coherency message is determined at the select coherency agent based on the address translation. The coherency message and a coherency domain identifier of the select coherency domain are provided by the select coherency agent to a coherency interconnect for distribution to at least one of the plurality of coherency agents based on the coherency domain identifier. | 01-15-2009 |
| 20090037666 | CACHE LOCKING DEVICE AND METHODS THEREOF - A method and device for locking a cache line of a cache is disclosed. The method includes automatically changing a state of a cache line from a valid locked state to an invalid locked state in response to receiving an indication that a memory location external to the cache and corresponding to the cache line is associated with an access request by a processor or other data access module. Thus, the locked state of a cache line is maintained even after data in the locked cache line is invalidated. By maintaining the invalid locked state, the cache line is not available for reallocation by the cache. This allows locked cache lines that become invalidated to remain locked without additional software overhead to periodically determine whether the lock has been lost due to invalidation of the cache line. | 02-05-2009 |
| 20090249048 | BRANCH TARGET BUFFER ADDRESSING IN A DATA PROCESSOR - A data processing system includes a branch target buffer (BTB) including a plurality of entries, each entry comprising a tag portion and a long branch indicator. The system also includes segment target address storage circuitry which stores a plurality of segment target addresses, index storage circuitry which stores a plurality of indices for indexing into the segment target address storage circuitry, and control circuitry which receives an instruction address and determines whether the instruction address matches a valid entry in the BTB. When the instruction address matches a valid entry in the BTB and the long branch indicator of the valid entry indicates a long branch, the index storage circuitry provides a selected index of the plurality of indices selected by the received instruction address. In response to the selected index, the segment target address storage circuitry provides a selected segment target address as a higher order target address portion. | 10-01-2009 |
| 20090249302 | Method and Apparatus to Trace and Correlate Data Trace and Instruction Trace for Out-of-Order Processors - In a data processing system, a marked bit is used to identify a data access instruction throughout the pipeline to indicate that the instruction meets user-specified criteria (e.g., a meets a data address range of interest). Based on the marked bit, an in-order program correlation message is generated which indicates when the data access instruction occurs relative to the instruction stream. The marked bit is also used to generate an in-order data trace message. As a result, the trace streams including only data access instructions meeting user-specified criteria may be post-processed and correlated precisely. | 10-01-2009 |
