Abstract: A system for predicting multiple targets for a single branch includes: a branch target buffer that includes a previous next address for an instruction and that receives an indirect instruction address to provide a first branch target prediction; a first branch table for capturing local past target information of an indirect branch in an encoded form; a second branch table which is a correlation table for storing potential branch targets based on a local branch history and which provides a second branch target prediction when the first branch target prediction is not successful; an exclusion predictor for inhibiting updates of inefficient entries; and a multiplexer to select the predicted target as output.

Abstract: A system and method for cache replacement includes: augmenting each cache block in a cache region with a region hint indicating a temporal priority of the cache block; receiving an indication that a cache miss has occurred; and selecting for eviction the cache block comprising the region hint indicating a low temporal priority.

Abstract: Methods and apparatus to dynamically insert prefetch instructions are disclosed. In an example method, one or more samples associated with cache misses are identified from a performance monitoring unit in a processor system. Based on sample information associated with the one or more samples, delinquent information is generated. To dynamically insert one or more prefetch instructions, a prefetch point is identified based on the delinquent information.

Abstract: A system and method for cache replacement includes: augmenting each cache block in a cache region with a region hint indicating a temporal priority of the cache block; receiving an indication that a cache miss has occurred; and selecting for eviction the cache block comprising the region hint indicating a low temporal priority.

Abstract: A computer-implemented method of cache replacement includes steps of: determining whether each cache block in a cache memory is a read or a write block; augmenting metadata associated with each cache block with an indicator of the type of access; receiving an access request resulting in a cache miss, the cache miss indicating that a cache block will need to be replaced; examining the indicator in the metadata of each cache block for determining a probability that said cache block will be replaced; and selecting for replacement the cache block with the highest probability of replacement.

Abstract: A system for predicting multiple targets for a single branch includes: a branch target buffer that includes a previous next address for an instruction and that receives an indirect instruction address to provide a first branch target prediction; a first branch table for capturing local past target information of an indirect branch in an encoded form; a second branch table which is a correlation table for storing potential branch targets based on a local branch history and which provides a second branch target prediction when the first branch target prediction is not successful; an exclusion predictor for inhibiting updates of inefficient entries; and a multiplexer to select the predicted target as output.

Abstract: An information processing system includes a branch target buffer (BTB) comprising the last next address for the instruction and for receiving an indirect instruction address and providing a BTB predicted target; and next branch target table (NBTT) for storing potential branch targets based on a history of the branch and for providing an NBTT when the a BTB predicted target is not successful. In another embodiment a system comprising a plurality of branch prediction resources dynamically predicts the best resource appropriate for a branch. The method includes predicting a target branch for an indirect instruction address using a resource chosen among the plurality of branch prediction resources; and selectively inhibiting updates of the branch prediction resources whose prediction accuracy does not meet a threshold.

Abstract: Methods and apparatus to insert prefetch instructions based on garbage collector analysis and compiler analysis are disclosed. In an example method, one or more batches of samples associated with cache misses from a performance monitoring unit in a processor system are received. One or more samples from the one or more batches of samples based on delinquent information are selected. A performance impact indicator associated with the one or more samples is generated. Based on the performance indicator, at least one of a garbage collector analysis and a compiler analysis is initiated to identify one or more delinquent paths. Based on the at least one of the garbage collector analysis and the compiler analysis, one or more prefetch points to insert prefetch instructions are identified.

Abstract: This invention describes a method and several variants for compiling programs or components of programs in a mixed static and dynamic environment, so as to reduce the amount of time and memory spent in run-time compilation, or to exercise greater control over testing of the executable code for the program, or both. The invention involves generating persistent code images prior to program execution based on static compilation or dynamic compilation from a previous run, and then, adapting those images during program execution. We describe a method for generating auxiliary information in addition to the executable code that is recorded in the persistent code image. Further, we describe a method for checking the validity of those code images, adapting those images to the new execution context, and generating new executable code to respond to dynamic events, during program execution.

Abstract: A method and several variants for using information about the scope of access of objects acted upon by mutual exclusion, or mutex, locks to transform a computer program by eliminating locking operations from the program or simplifying the locking operations, while strictly performing the semantics of the original program. In particular, if it can be determined by a compiler that the object locked can only be accessed by a single thread it is not necessary to perform the “acquire” or “release” part of the locking operation, and only its side effects must be performed. Likewise, if it can be determined that the side effects of a locking operation acting on a variable which is locked in multiple threads are not needed, then only the locking operation, and not the side effects, needs to be performed.

Abstract: Information is computed about the reachability relationships among objects and pointers to enable transformation of a computer program for optimizing the creation and destruction of objects, while strictly performing the semantics of the original program. An interprocedural analysis is used to determine whether an object that is allocated on the heap during the execution of a procedure is not reachable from any global variable, parameter, or the return value of the procedure after it returns. If so, that object can be allocated on the stack frame of the procedure in which it is otherwise heap-allocated. This simplifies the memory allocation and deallocation operations, as allocation on the stack can be done more efficiently than allocation on the heap, and objects allocated on the stack frame of a procedure are automatically deallocated when the procedure returns, without incurring the overhead of garbage collection.