International Business Machines Corporation

Alexander Erik Mericas

Venkat Rajeev Indukuru

William John Starke

Joel Roger Davidson

Robert Dominick Mirabella

Ra'ed Mohammad Al-Omari

Michael Stephen Floyd

Hung Qui Le

Kevin Franklin Reick

Joel M Tendler

Toshihiko Kurihara

Michitaka Okuno

Masahiro Tokoro

Judith E K Laurens

John Fred Spannaus

Satish Kumar Sadasivam

Brian R Konigsburg

Benjamin W Stolt

Madhavi Gopal Valluri

Zhong Liang Wang

Daniel Owen

Judith K Laurens

Dung Q Nguyen

Balaram Sinharoy

James Allan Kahle

Giles R Frazier

Paul Joseph Jordan

Ronald P Hall

Wen-Tzer Thomas Chen, Jr

Mauricio J Serrano

Thomas Pflueger

Tilman Gloekler

Mysore Sathyanarayana Srinivas

Richard William Doing

James Wilson Bishop

Philip Lee Vitale

John Edward Derrick

Matthias Fertig

Ralph C Koester

Douglas Robert Logan

Dennis G Gregoire

Daniel Becker

Hideki Mitsubayashi

Hung Oui Le

A performance monitor including a saturating counter provides a relative measure of event frequency without requiring a minimum polling rate or periodic reset to avoid or account for counter overflow. The saturating counter is incremented upon detection of an event and decremented if an event is not detected within a predetermined period. The period of detecting may be programmable and may be determined by real time clock, processor or instruction cycles. Multiple event types may be selected from for detection and input to a single counter, or alternatively multiple event counters may be provided for various event types. The saturating counter may additionally be periodically reset in a selected operating mode, in combination with the decrementing action performed on the counter.

Processor core having a saturating event counter for making performance measurements

During a pipeline stall in a processor, until a next to complete instruction group completes, a monitoring unit receives, from a completion unit of a processor, a next to finish indicator indicating the finish of an oldest previously unfinished instruction from among a plurality of instructions of a next to complete instruction group. The monitoring unit receives, from functional units of the processor, finish reports including completion reasons for separate instructions. The monitoring unit determines at least one stall reason from among multiple stall reasons for the oldest instruction from a selection of completion reasons from a selection of finish reports aligned with the next to finish indicator from among the finish reports. Once the monitoring unit receives a complete indicator from the completion unit, indicating the completion of the next to complete instruction group, the monitoring unit stores each determined stall reason aligned with each next to finish indicator in memory.

Determining each stall reason for each stalled instruction within a group of instructions during a pipeline stall

Methods, systems, and computer program products may provide delay-identification in data processing systems. An apparatus may include a delay-identification unit having a delay counter, a threshold register, a delay register, and a delay detector. The delay detector may be configured to start the delay counter in response to detecting that one group of instructions is delayed, and stop the delay counter in response to detecting that the one group of instructions is no longer delayed. The delay detector may additionally be configured to compare the number of cycles counted by the delay counter with a threshold number of cycles in the threshold register, and store at least one effective address of one of the instructions of the one group of instructions when the number of cycles counted by the delay counter is greater than the threshold number of cycles stored in the threshold register.

Prioritizing instructions based on the number of delay cycles

A processor performance profiler is enabled to for identify specific instructions causing performance issues within a program being executed by a microprocessor through random sampling to find the worst-case offenders of a particular event type such as a cache miss or a branch mis-prediction. Tracking all instructions causing a particular event generates large data logs, creates performance penalties, and makes code analysis more difficult. However, by identifying and tracking the worst offenders within a random sample of events without having to hash all events results in smaller memory requirements for the performance profiler, lower performance impact while profiling, and decreased complexity to analyze the program to identify major performance issues, which, in turn, enables better optimization of the program in shorter developer time.

Autonomic hotspot profiling using paired performance sampling

A computing device identifies a load instruction and store instruction pair that causes a load-hit-store conflict. A processor tags a first load instruction that instructs the processor to load a first data set from memory. The processor stores an address at which the first load instruction is located in memory in a special purpose register. The processor determines where the first load instruction has a load-hit-store conflict with a first store instruction. If the processor determines the first load instruction has a load-hit store conflict with the first store instruction, the processor stores an address at which the first data set is located in memory in a second special purpose register, tags the first data set being stored by the first store instruction, stores an address at which the first store instruction is located in memory in a third special purpose register and increases a conflict counter.

Identifying load-hit-store conflicts

An information handling system includes a processor that executes multiple instructions or instruction threads within a software application program. The information handling system includes operating system software that manages processor system hardware and software in a multi-tasking environment. In one embodiment, the operating system manages instruction completion stall analysis software to determine the cause or causes of instruction stalls. In another embodiment, the stall analysis software cooperates with the operating system software to store instruction completion stall event data on a per instruction basis while the application program executes. The operating system software may cooperate with the stall analysis software to store instruction completion stall data in memory for later manipulation by system users or other software.

Method and apparatus for instruction completion stall identification in an information handling system

During a pipeline stall in an out of order processor, until a next to complete instruction group completes, a monitoring unit receives, from a completion unit of a processor, a next to finish indicator indicating the finish of an oldest previously unfinished instruction from among a plurality of instructions of a next to complete instruction group. The monitoring unit receives, from a plurality of functional units of the processor, a plurality of finish reports including completion reasons for a plurality of separate instructions. The monitoring unit determines at least one stall reason from among multiple stall reasons for the oldest instruction from a selection of completion reasons from a selection of finish reports aligned with the next to finish indicator from among the plurality of finish reports. Once the monitoring unit receives a complete indicator from the completion unit, indicating the completion of the next to complete instruction group, the monitoring unit stores each determined stall reason aligned with each next to finish indicator in memory.

A method and data processing system for managing running of instructions in a program. A processor of the data processing system receives a monitoring instruction of a monitoring unit. The processor determines if at least one secondary thread of a set of secondary threads is available for use as an assist thread. The processor selects the at least one secondary thread from the set of secondary threads to become the assist thread in response to a determination that the at least one secondary thread of the set of secondary threads is available for use as an assist thread. The processor changes profiling of running of instructions in the program from the main thread to the assist thread.

Hardware assist thread for dynamic performance profiling

Occurrences of a particular event in an electronic device are counted by incrementing an event counter each time a variable number of the particular events have occurred, and automatically increasing that variable number as the total count increases. The variable number (prescale value) can increase geometrically according to a programmable counter base each time the count mantissa overflows. The event counter thereby provides hardware-implemented automatic prescaling while significantly reducing the number of interface bits required to support very large count ranges, and retaining high accuracy at very large event counts.

Floating-point event counters with automatic prescaling

An information handling system includes a memory, a processor, and an instruction tracking unit. The processor executes program code and, while the program code executes, the instruction tracking unit decodes a multi-purpose no-op instruction within the program code. In turn, the instruction tracking unit sends an interrupt to the processor, which invokes a profiling module to collect and store profiling data in a profiling buffer.

General purpose emit for use in value profiling

A method, computer program product, and data processing system for collecting metrics regarding completion stalls in an out-of-order superscalar processor with branch prediction is disclosed. A preferred embodiment of the present invention selectively samples particular instructions (or classes of instructions). Each selected instruction, as it passes through the processor datapath, is marked (tagged) for monitoring by a performance monitoring unit. The progress of marked instructions is monitored by the performance monitoring unit, and various stall counters are triggered by the progress of the marked instructions and the instruction groups they form a part of. The stall counters count cycles to give an indication of when certain delays associated with particular instructions occur and how serious the delays are.

Quantifying completion stalls using instruction sampling

A system and associated method for distributing signals with efficiency over a microprocessor. A performance monitoring unit (PMU) sends configuration signals to a unit to monitor an event occurring on the unit. The unit is attached to a configuration bus and an event bus that are daisy-chained from PMU to other units in the microprocessor. The configuration bus transmits configuration signals from the PMU to the unit to set the unit to report the event. The unit sends event signals to the PMU through the event bus. The unit is configured upon receiving configuration signals comprising a base address of a bus ramp of the unit. A number of units and a number of events for monitoring is flexibly selected by adjusting a length of bit fields within configuration signals.

System and method for distributing signal with efficiency over microprocessor

An apparatus, and computer program product are disclosed in a processor for dynamically, during runtime, allocating memory for in-memory hardware tracing. The processor is included within a data processing system. The processor includes multiple processing units that are coupled together utilizing a system bus. The processing units include a memory controller that controls a system memory. A particular size of the system memory is determined that is needed for storing trace data. A hardware trace facility requests, dynamically after the data processing system has completed booting, the particular size of the system memory to be allocated to the hardware trace facility for storing trace data that is captured by the hardware trace facility. The firmware selects particular locations within the system memory. All of the particular locations together are the particular size. The firmware allocates the particular locations for use exclusively by the hardware trace facility.

Method, apparatus, and computer program product in a processor for dynamically during runtime allocating memory for in-memory hardware tracing

Growing community of inventors

Austin, TX, United States of America

Alexander Erik Mericas