The report contains two levels of information:

•Exact counts of CPU metrics summed across the entire run of an application

•Sampled IPs that are associated with particular locations in the application

The default for the fprof measurement is to take a sample every 500,000 +/- 25,000 CPU cycles. (CPU_CYCLES is the event.)

You use the -s(--sampling-spec) option to change both the event being sampled and the interval. For example, if you want to keep the same variation and sample event and simply change the interval, you can use the following option: -s 100000. The sampling rate, variation, and event used are listed in the Sampling Specification part of the report.

Command-line options allow you to control the amount of data reported, how the data is sorted, and the number of statements and instructions reported for each sampled program location.

Example Command Line for Text Report

$ caliper fprof -o reports/sample.txt ./wordplay thequickbrownfox

Example Command Line for CSV Report

$ caliper fprof --csv csvout ./wordplay thequickbrownfox

fprof Metrics Summed for Entire Run

This section describes the metrics summed over the entire run of your application under HP Caliper.

Metrics for Integrity Servers Itanium 2 Systems

CPU_CYCLES

Number of elapsed processor cycles.

BACK_END_BUBBLE.ALL

Full pipe bubbles in main pipe.

BE_EXE_BUBBLE.GRALL

Full pipe bubbles in main pipe due to execution unit stalls.

% of Cycles Lost Due to Stalls

Percentage of cycles lost due to stalls.

% of Cycles Stalled Due to GR/GR

Percentage of cycles stalled due to general register/general

or GR/Load Dependency

register or general register/load dependency.

Metrics for Integrity Servers Dual-Core Itanium 2 and Itanium 9300 Quad-Core Processor Systems

BACK_END_BUBBLE.FE

Full pipe bubbles in main pipe due to front end. This is the

 

number of cycles lost (stall cycles) due to instruction cache,

 

ITLB, and branch execution stalls.

BE_EXE_BUBBLE.ALL

Full pipe bubbles in main pipe due to execution unit stalls.

 

This is the number of cycles lost (stall cycles) due to stalls

 

caused by the execution unit.

BE_EXE_BUBBLE.FRALL

Full Pipe Bubbles in Main Pipe due to FR/FR or FR/load

 

dependency stalls. This is the number of cycles lost (stall

 

cycles) due to FR/FR or FR/load dependency.

BE_EXE_BUBBLE.GRALL

Full pipe bubbles in main pipe due to general

 

register/general register or general register/load

 

dependency stalls. This is the number of cycles lost (stall

 

cycles) due to general register/general register or general

 

register/load dependency.

BE_EXE_BUBBLE.GRGR

Full Pipe Bubbles in Main Pipe due to GR/GR dependency

 

stalls. This is the number of cycles lost (stall cycles) due to

 

GR/GR dependency stalls.

BE_FLUSH_BUBBLE.ALL

Full Pipe Bubbles in Main Pipe due to pipeline flushes. This

 

is the number of cycles lost (stall cycles) due to branch

 

misprediction or exception/interruption flush.

fprof Measurement Report Description 201