coz - profiler running experiements on multithreaded code
Contents
Copyright
public domain
0.1 2016-08-15 COZ(1)
Description
Coz is a new kind of profiler that unlocks optimization opportunities missed by traditional profilers.
Coz employs a novel technique we call causalprofiling that measures optimization potential. This
measurement matches developers' assumptions about profilers: that optimizing highly-ranked code will have
the greatest impact on performance. Causal profiling measures optimization potential for serial,
parallel, and asynchronous programs without instrumentation of special handling for library calls and
concurrency primitives. Instead, a causal profiler uses performance experiments to predict the effect of
optimizations. This allows the profiler to establish causality: "optimizing function X will have effect
Y," exactly the measurement developers had assumed they were getting all along.
Name
coz - profiler running experiements on multithreaded code
Options
-h, --help
show this help message and exit
--binary-scope<filepattern>, -b<filepattern>
Profile matching executables. Use '%' as a wildcard, or 'MAIN' to include the main executable
(default=MAIN)
--source-scope<filepattern>, -s<filepattern>
Profile matching source files. Use '%' as a wildcard. (default=%)
--progress<sourcefile>:<linenumber>,-p<sourcefile>:<linenumber>
[NOT SUPPORTED] Add a sampling-based progress point
--output<profileoutput>, -o<profileoutput>
Profiler output (default=`profile.coz`)
--end-to-end
Run a single performance experiment per-execution
--fixed-line<sourcefile>:<linenumber>
Evaluate optimizations of a specific source line
--fixed-speedup<speedup>(0-100)
Evaluate optimizations of a specific amount
See Also
• http://coz-profiler.org
• mangdbSynopsis
coz run [profiling options] --- <command> [args]
coz plot [-h]
