Abstract: Handling corrupted background data in an out of order processing environment. Modified data is stored on a byte of a word having at least one byte of background data. A byte valid vector and a byte store bit are added to the word. Parity checking is done on the word. If the word does not contain corrupted background date, the word is propagated to the next level of cache. If the word contains corrupted background data, a copy of the word is fetched from a next level of cache that is ECC protected, the byte having the modified data is extracted from the word and swapped for the corresponding byte in the word copy. The word copy is then written into the next level of cache that is ECC protected.

Abstract: A mechanism is provided for avoiding cross-interrogates for a streaming data optimized level one cache. The mechanism adds a set of dedicated registers, referred to as “copex registers,” to track ownership of the cache lines that the co-processor's L1 cache holds exclusive. The mechanism extends the cache directory of the L2 cache by a bit that identifies exclusive ownership of a cache line in the co-processor cache. The co-processor continuously provides an indication of which copex registers are valid. On any action that requires a directory lookup in the L2 cache, the mechanism compares the valid copex registers against the lookup address in parallel to the directory lookup. The mechanism considers the “exclusive ownership in co-processor” bit in the directory valid only if the cache line is also currently in a valid copex register.

Abstract: An architecture for collecting performance data in a processor, that includes: a trace read control unit and a trace data collect unit, each unit coupled to a plurality of trace array and multiplex units for providing performance data, the coupling accomplished by a trace read control bus, a data select bus, a trace row address bus and a data return bus; wherein each of the trace array and multiplex units receives a trace read signal and provides data including trace data and the trace read signal to the trace data collect unit. A method and a computer program product are provided.

Abstract: A system for 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: Data caching for use in a computer system including a lower cache memory and a higher cache memory. The higher cache memory receives a fetch request. It is then determined by the higher cache memory the state of the entry to be replaced next. If the state of the entry to be replaced next indicates that the entry is exclusively owned or modified, the state of the entry to be replaced next is changed such that a following cache access is processed at a higher speed compared to an access processed if the state would stay unchanged.

Abstract: The present invention relates to a method, computer program product and system for generating a sample signal from differing clock domain boundaries. The system comprises a cycle base component, a sample offset component being configured to receive a time-based sample pulse signal, and logic to generate a sample pulse. The sample pulse generation logic is configured to receive a time-based sample pulse signal, a free running counter value, a sample offset counter value, and deliver a sample pulse signal.

Abstract: A system, method and computer program product for event-based sampling to monitor computer system performance are provided. The system includes a sample buffer to store a sample of instrumentation data, where the instrumentation data enables measurement of computer system performance. The system also includes a sample segment selector to isolate a segment of the sample of instrumentation data as an event. The system further includes an instrumentation counter counting in response to a combination of the event and a sample pulse, and asserting a sample interrupt indicating that the sample of instrumentation data is ready to logout from the sample buffer.

Abstract: A system, method and computer program product for sampling computer system performance data are provided. The system includes a sample buffer to store instrumentation data while capturing trace data in a trace array, where the instrumentation data enables measurement of computer system performance. The system further includes a sample interrupt generator to assert a sample interrupt indicating that the instrumentation data is available to read. The sample interrupt is asserted in response to storing the instrumentation data in the sample buffer.

Abstract: A method and apparatus for changing a clock frequency in a system (10) comprising a plurality of synchronous integrated circuit chips (12, 14, 16), and a circuit (20) for implementing the frequency change. The method includes: detecting a change in processing requirements in one of the plurality of synchronous integrated circuit chips; notifying the plurality of synchronous integrated circuit chips that a clock frequency change is to occur; achieving a quiescent bus state in each of the plurality of synchronous integrated circuit chips; notifying the plurality of synchronous integrated circuit chips that the clock frequency change can occur; and changing the clock frequency of the plurality of integrated circuit chips.

Abstract: The illustrative embodiments provide accurate time-based counters for scaling operating frequencies of microprocessors. A time-based counter circuit configuration in which a fixed frequency clock is derived from a PLL of the clock generation circuit of the microprocessor and is used to feed the external and internal timebase logic as well as a timebase accumulator counter. The timebase accumulator counter accumulates the tick events from the timebase logic between two core clocks. The accumulated value is transferred to the core clock domain on every clock edge of a scalable clock and the accumulator is then reset. Because the accumulated ticks are transferred to the core clock domain before the accumulator is reset, no ticks are ever lost using the circuitry of the illustrative embodiment.

Abstract: A mechanism provides accurate time-based counters for scaling operating frequencies of microprocessors. The mechanism makes use of a time-based counter circuit configuration in which a fixed frequency clock is derived from a PLL of the clock generation circuit of the microprocessor and is used to feed the external and internal timebase logic as well as a timebase accumulator counter. The timebase accumulator counter accumulates the tick events from the timebase logic between two core clocks. The accumulated value is transferred to the core clock domain on every clock edge of a scalable clock and the accumulator is then reset. Because the accumulated ticks are transferred to the core clock domain before the accumulator is reset, no ticks are ever lost using the circuitry of the illustrative embodiment.

Abstract: An architecture for collecting performance data in a processor, that includes: a trace read control unit and a trace data collect unit, each unit coupled to a plurality of trace array and multiplex units for providing performance data, the coupling accomplished by a trace read control bus, a data select bus, a trace row address bus and a data return bus; wherein each of the trace array and multiplex units receives a trace read signal and provides data including trace data and the trace read signal to the trace data collect unit. A method and a computer program product are provided.

Abstract: A system, method and computer program product for event-based sampling to monitor computer system performance are provided. The system includes a sample buffer to store a sample of instrumentation data, where the instrumentation data enables measurement of computer system performance. The system also includes a sample segment selector to isolate a segment of the sample of instrumentation data as an event. The system further includes an instrumentation counter counting in response to a combination of the event and a sample pulse, and asserting a sample interrupt indicating that the sample of instrumentation data is ready to logout from the sample buffer.

Abstract: The present invention relates to a method, computer program product and system for generating a sample signal from differing clock domain boundaries. The system comprises a cycle base component, a sample offset component being configured to receive a time-based sample pulse signal, and logic to generate a sample pulse. The sample pulse generation logic is configured to receive a time-based sample pulse signal, a free running counter value, a sample offset counter value, and deliver a sample pulse signal.

Abstract: A system, method and computer program product for sampling computer system performance data are provided. The system includes a sample buffer to store instrumentation data while capturing trace data in a trace array, where the instrumentation data enables measurement of computer system performance. The system further includes a sample interrupt generator to assert a sample interrupt indicating that the instrumentation data is available to read. The sample interrupt is asserted in response to storing the instrumentation data in the sample buffer.

Abstract: The illustrative embodiments provide accurate time-based counters for scaling operating frequencies of microprocessors. A time-based counter circuit configuration in which a fixed frequency clock is derived from a PLL of the clock generation circuit of the microprocessor and is used to feed the external and internal timebase logic as well as a timebase accumulator counter. The timebase accumulator counter accumulates the tick events from the timebase logic between two core clocks. The accumulated value is transferred to the core clock domain on every clock edge of a scalable clock and the accumulator is then reset. Because the accumulated ticks are transferred to the core clock domain before the accumulator is reset, no ticks are ever lost using the circuitry of the illustrative embodiment.

Abstract: A method and apparatus for changing a clock frequency in a system (10) comprising a plurality of synchronous integrated circuit chips (12, 14, 16), and a circuit (20) for implementing the frequency change. The method includes: detecting a change in processing requirements in one of the plurality of synchronous integrated circuit chips; notifying the plurality of synchronous integrated circuit chips that a clock frequency change is to occur; achieving a quiescent bus state in each of the plurality of synchronous integrated circuit chips; notifying the plurality of synchronous integrated circuit chips that the clock frequency change can occur; and changing the clock frequency of the plurality of integrated circuit chips.

Abstract: The illustrative embodiments provide a system, apparatus and method for providing accurate time-based counters for scaling operating frequencies of microprocessors. The system, apparatus and method make use of a time-based counter circuit configuration in which a fixed frequency clock is derived from a PLL of the clock generation circuit of the microprocessor and is used to feed the external and internal timebase logic as well as a timebase accumulator counter. The timebase accumulator counter accumulates the tick events from the timebase logic between two core clocks. The accumulated value is transferred to the core clock domain on every clock edge of a scalable clock and the accumulator is then reset. Because the accumulated ticks are transferred to the core clock domain before the accumulator is reset, no ticks are ever lost using the circuitry of the illustrative embodiment.