Abstract: A hierarchical cache structure comprises at least one higher level cache comprising a unified cache array for data and instructions and at least two lower level caches, each split in an instruction cache and a data cache. An instruction cache and a data cache of a split second level cache are connected to a third level cache; and an instruction cache of a split first level cache is connected to the instruction cache of the split second level cache, and a data cache of the split first level cache is connected to the instruction cache and the data cache of the split second level cache.

Abstract: A tool for determining eviction of store cache entries based on store pressure. The tool determines, by one or more computer processors, a count value for one or more new store cache entry allocations. The tool determines, by one or more computer processors, whether a new store cache entry allocation limit is exceeded. Responsive to determining the new store cache entry allocation limit is exceeded, the tool determines, by one or more computer processors, an allocation value for one or more existing store cache entries, the allocation value indicating an allocation class for each of the one or more existing store cache entries. The tool determines, by one or more computer processors based, at least in part, on the allocation value for the one or more existing store cache entries, at least one allocation class for eviction. The tool program determines, by one or more computer processors, an eviction request setting for evicting the one or more existing store cache entries.

Abstract: Cache miss rates for threads operating in a simultaneous multi-threading computer processing environment can be estimated. The single thread rates can be estimated by monitoring a shared directory for cache misses for a first thread. Memory access requests can be routed to metering cache directories associated with the particular thread. Single thread misses to the shared directory and single thread misses to the associated metering cache directory are monitored and a performance indication is determined by comparing the cache misses with the thread misses. The directory in the associated metering cache is rotated, and a second sharing performance indication is determined.

Abstract: Cache miss rates for threads operating in a simultaneous multi-threading computer processing environment can be estimated. The single thread rates can be estimated by monitoring a shared directory for cache misses for a first thread. Memory access requests can be routed to metering cache directories associated with the particular thread. Single thread misses to the shared directory and single thread misses to the associated metering cache directory are monitored and a performance indication is determined by comparing the cache misses with the thread misses. The directory in the associated metering cache is rotated, and a second sharing performance indication is determined.

Abstract: A method to merge one or more non-transactional stores and one or more thread-specific transactional stores into one or more cache line templates in a store buffer in a store cache. The method receives a thread-specific non-transactional store address and a first data, maps the store address to a first cache line template, and merges the first data into the first cache line template, according to a store policy. The method further receives a thread-specific transactional store address and a second data, maps the thread-specific store address into a second cache line template, according to a store policy. The method further writes back a copy of a cache line template to a cache and invalidates a third cache line template, which frees the third cache line template from a store address mapping.

Abstract: A method to merge one or more non-transactional stores and one or more thread-specific transactional stores into one or more cache line templates in a store buffer in a store cache. The method receives a thread-specific non-transactional store address and a first data, maps the store address to a first cache line template, and merges the first data into the first cache line template, according to a store policy. The method further receives a thread-specific transactional store address and a second data, maps the thread-specific store address into a second cache line template, according to a store policy. The method further writes back a copy of a cache line template to a cache and invalidates a third cache line template, which frees the third cache line template from a store address mapping.

Abstract: A hierarchical cache structure includes at least one real indexed higher level cache with a directory and a unified cache array for data and instructions, and at least two lower level caches, each split in an instruction cache and a data cache. An instruction cache of a split real indexed second level cache includes a directory and a corresponding cache array connected to the real indexed third level cache. A data cache of the split second level cache includes a directory connected to the third level cache. An instruction cache of a split virtually indexed first level cache is connected to the second level instruction cache. A cache array of a data cache of the first level cache is connected to the cache array of the second level instruction cache and to the cache array of the third level cache. A directory of the first level data cache is connected to the second level instruction cache directory and to the third level cache directory.

Abstract: Cache miss rates for threads operating in a simultaneous multi-threading computer processing environment can be estimated. The single thread rates can be estimated by monitoring a shared directory for cache misses for a first thread. Memory access requests can be routed to metering cache directories associated with the particular thread. Single thread misses to the shared directory and single thread misses to the associated metering cache directory are monitored and a performance indication is determined by comparing the cache misses with the thread misses. The directory in the associated metering cache is rotated, and a second sharing performance indication is determined.

Abstract: Cache miss rates for threads operating in a simultaneous multi-threading computer processing environment can be estimated. The single thread rates can be estimated by monitoring a shared directory for cache misses for a first thread. Memory access requests can be routed to metering cache directories associated with the particular thread. Single thread misses to the shared directory and single thread misses to the associated metering cache directory are monitored and a performance indication is determined by comparing the cache misses with the thread misses. The directory in the associated metering cache is rotated, and a second sharing performance indication is determined.

Abstract: Data operand fetching control includes a computer processor that includes a control unit for determining memory access operations. The control unit is configured to perform a method. The method includes calculating a summation weight value for each instruction in a pipeline, the summation weight value calculated as a function of branch uncertainty and a pendency in which the instruction resides in the pipeline relative to other instructions in the pipeline. The method also includes mapping the summation weight value of a selected instruction that is attempting to access system memory to a memory access control, each memory access control specifying a manner of handling data fetching operations. The method further includes performing a memory access operation for the selected instruction based upon the mapping.

Abstract: A hierarchical cache structure comprises at least one higher level cache comprising a unified cache array for data and instructions and at least two lower level caches, each split in an instruction cache and a data cache. An instruction cache and a data cache of a split second level cache are connected to a third level cache; and an instruction cache of a split first level cache is connected to the instruction cache of the split second level cache, and a data cache of the split first level cache is connected to the instruction cache and the data cache of the split second level cache.

Abstract: Systems and methods for providing data from a cache memory to requestors includes a number of cache memory levels arranged in a hierarchy. The method includes receiving a request for fetching data from the cache memory and determining one or more addresses in a cache memory level which is one level higher than a current cache memory level using one or more prediction algorithms. Further, the method includes pre-fetching the one or more addresses from the high cache memory level and determining if the data is available in the addresses. If data is available in the one or more addresses then data is fetched from the high cache level, else addresses of a next level which is higher than the high cache memory level are determined and pre-fetched. Furthermore, the method includes providing the fetched data to the requestor.

Abstract: A method to merge one or more non-transactional stores and one or more thread-specific transactional stores into one or more cache line templates in a store buffer in a store cache. The method receives a thread-specific non-transactional store address and a first data, maps the store address to a first cache line template, and merges the first data into the first cache line template, according to a store policy. The method further receives a thread-specific transactional store address and a second data, maps the thread-specific store address into a second cache line template, according to a store policy. The method further writes back a copy of a cache line template to a cache and invalidates a third cache line template, which frees the third cache line template from a store address mapping.

Abstract: A method to merge one or more non-transactional stores and one or more thread-specific transactional stores into one or more cache line templates in a store buffer in a store cache. The method receives a thread-specific non-transactional store address and a first data, maps the store address to a first cache line template, and merges the first data into the first cache line template, according to a store policy. The method further receives a thread-specific transactional store address and a second data, maps the thread-specific store address into a second cache line template, according to a store policy. The method further writes back a copy of a cache line template to a cache and invalidates a third cache line template, which frees the third cache line template from a store address mapping.

Abstract: A hierarchical cache structure comprises at least one higher level cache comprising a unified cache array for data and instructions and at least two lower level caches, each split in an instruction cache and a data cache. An instruction cache and a data cache of a split second level cache are connected to a third level cache; and an instruction cache of a split first level cache is connected to the instruction cache of the split second level cache, and a data cache of the split first level cache is connected to the instruction cache and the data cache of the split second level cache.

Abstract: A mechanism for simultaneous multithreading is provided. Responsive to performing a store instruction for a given thread of threads on a processor core and responsive to the core having ownership of a cache line in a cache, an entry of the store instruction is placed in a given store queue belonging to the given thread. The entry for the store instruction has a starting memory address and an ending memory address on the cache line. The starting memory addresses through ending memory addresses of load queues of the threads are compared on a byte-per-byte basis against the starting through ending memory address of the store instruction. Responsive to one memory address byte in the starting through ending memory addresses in the load queues overlapping with a memory address byte in the starting through ending memory address of the store instruction, the threads having the one memory address byte is flushed.

Abstract: A mechanism for simultaneous multithreading is provided. Responsive to performing a store instruction for a given thread of threads on a processor core and responsive to the core having ownership of a cache line in a cache, an entry of the store instruction is placed in a given store queue belonging to the given thread. The entry for the store instruction has a starting memory address and an ending memory address on the cache line. The starting memory addresses through ending memory addresses of load queues of the threads are compared on a byte-per-byte basis against the starting through ending memory address of the store instruction. Responsive to one memory address byte in the starting through ending memory addresses in the load queues overlapping with a memory address byte in the starting through ending memory address of the store instruction, the threads having the one memory address byte is flushed.

Abstract: Fetching a cache line into a plurality of caches of a multilevel cache system. The multilevel cache system includes at least a first cache, a second cache on a next higher level and a memory, the first cache being arranged to hold a subset of information of the second cache, the second cache being arranged to hold a subset of information of a next higher level cache or memory if no higher level cache exists. A fetch request is sent from one cache to the next cache in the multilevel cache system. The cache line is fetched in a particular state into one of the caches, and in another state into at least one of the other caches.

Abstract: A tool for determining eviction of store cache entries based on store pressure. The tool determines, by one or more computer processors, a count value for one or more new store cache entry allocations. The tool determines whether a new store cache entry allocation limit is exceeded. Responsive to determining the new store cache entry allocation limit is exceeded, the tool determines an allocation value for one or more existing store cache entries, the allocation value indicating an allocation class for each of the one or more existing store cache entries. The tool determines based, at least in part, on the allocation value for the one or more existing store cache entries, at least one allocation class for eviction. The tool program determines an eviction request setting for evicting the one or more existing store cache entries.

Abstract: A hierarchical cache structure includes at least one real indexed higher level cache with a directory and a unified cache array for data and instructions, and at least two lower level caches, each split in an instruction cache and a data cache. An instruction cache of a split real indexed second level cache includes a directory and a corresponding cache array connected to the real indexed third level cache. A data cache of the split second level cache includes a directory connected to the third level cache. An instruction cache of a split virtually indexed first level cache is connected to the second level instruction cache. A cache array of a data cache of the first level cache is connected to the cache array of the second level instruction cache and to the cache array of the third level cache. A directory of the first level data cache is connected to the second level instruction cache directory and to the third level cache directory.