This section describes changes made to various command-line performance analysis tools. For more information, see the corresponding man pages for each command-line tool.
Data collection tools include the collect command, dbx collector command, and er_kernel command. Each of these tools is used to profile programs to collect data and create experiments that can be read by Performance Analyzer or er_print. All data collection tools have the following changes.
Tracing of Java garbage collection is automatically done for Java experiments.
For Java, the synctrace feature has been extended with an optional <scope> modifier; <scope> can be n for native API tracing, j for Java API tracing, or nj for tracing both APIs. nj is the default.
Hardware counters on Oracle SPARC and x86 (including Haswell-E/EP) have updated support and memoryspace views.
New hardware counter option –auto helps select appropriate profile rates.
New guidance on using SPARC hardware counters is available using collect –h and er_kernel–h.
Issues concerning dlopen(), dlmopen(), dlclose(), exit() and Exit() are fixed.
The threads created on Linux when CLONE_VM is specified will not be followed.
The limit on the maximum number of threads has been removed.
The collect utility is a tool you use to profile your application as it runs to collect data and create an experiment that can be read by Performance Analyzer or er_print.
In addition to the changes common to all data collection tools, the collect utility is changed in this release as follows:
Heap tracing on Java targets will be allowed, but it will only trace native allocations, not Java allocations
The –R argument is no longer recognized.
The dbx collector is a subcommand of the dbx debugger that you can use for performance data collection. See the collector(1) man page for more information.
In addition to the changes common to all data collection tools, the dbx collector command is changed in this release as follows:
Several bugs in the 12.4 version have been fixed.
The er_kernel command profiles the Oracle Solaris kernel and generates an experiment that you can examine in Performance Analyzer or er_print.
In addition to the changes common to all data collection tools, the er_kernel utility is changed as follows:
Process creation and termination are more accurately tracked.
The formatting of the er_kernel –h output has been improved.
See the er_kernel(1) man page for more information.
The er_print utility generates a plain-text version of the data views presented by the Performance Analyzer. The output is displayed on the standard output.
The er_print utility is changed in this release as follows:
The er_print utility is changed as follows:
Several bugs in the 12.4 version have been fixed.
The flags used when compiling with Oracle Developer Studio are shown in the source and disassembly reports.
Machine model information is reported in the experiment header.
The overview command has been implemented.
See the er_print(1)man page for more information.
The libcollector API has the following updates:
Static versions of libcollectorAPI.a and libfcollector.a are now available.
The man page has been rewritten to better describe the Java API. See libcollector(3).