hpcprof-mpi scalably analyzes call path profile performance measurements (in parallel) and attributes them to static source code structure. See hpctoolkit(1) for an overview of HPCToolkit.
hpcprof-mpi [options] measurement-group...
hpcprof-mpi analyzes call path profile performance measurements, attributes them to static source code structure, and generates an Experiment database for use with hpcviewer(1) . hpcprof-mpi is especially designed for analyzing and attributing measurements from large-scale executions.
hpcprof-mpi expects a list of measurement-groups, where a group is a call path profile directory or an individual profile file. For best results, two other options should be used: -I to provide search directories for source code and -S to provide source code structure from hpcstruct(1) . (Note that without any search-directory argument, hpcprof will only find source files that either (1) have absolute paths (and that still exist on the file system) or (2) are relative to the current working directory.)
Default values for an option's optional arguments are shown in {}.
Note: With multiple search-directory arguments, it may be the case that file f exists within more than one search directory. In this case, the ambiguity is resolved in favor of the search directory that appears first on the command line.
Use this option when a profile or binary contains references to files that have been relocated, such as might occur with a file system change.
hpctoolkit(1)
.
hpcprof(1)
.
hpcproftt(1)
.
hpcsummary(1)
.
Version: 5.2.1-r3669 of 2011/02/22.
Nathan Tallent
John Mellor-Crummey
Rice HPCToolkit Research Group
Email: hpctoolkit-forum =at= rice.edu
WWW: http://hpctoolkit.org.