\input texinfo @c -*-texinfo-*- @c for $sect (qw(NAME SYNOPSIS TARGET DESCRIPTION OPTIONS ENVIRONMENT FILES @c BUGS NOTES FOOTNOTES SEEALSO AUTHOR COPYRIGHT)) { @c ---------------------------------------------------------------------------- @c This is the Texinfo source file for the GPROFNG manual. @c @c Author: Ruud van der Pas @c ---------------------------------------------------------------------------- @c %**start of header @setfilename gprofng.info @settitle GNU gprofng @c -- Set the indent for the @example command to 1 space, not 5 --------------- @exampleindent 1 @c %**end of header @c -- Start a new chapter on a new, odd numbered, page ------------------------ @setchapternewpage odd @c -- Merge all index entries into the Concepts Index ------------------------- @syncodeindex fn cp @syncodeindex ky cp @syncodeindex pg cp @syncodeindex vr cp @c -- Macro definitions ------------------------------------------------------- @c @c Since only letters can be used, we use capitalization to distinguish @c different words. @c ---------------------------------------------------------------------------- @macro CollectApp{} @command{gprofng collect app} @end macro @macro DisplayHTML{} @command{gprofng display html} @end macro @macro DisplayText{} @command{gprofng display text} @end macro @macro Driver{} @command{gprofng} @end macro @macro ProductName{} gprofng @end macro @macro ToolName{} @command{gprofng} @end macro @macro IndexSubentry{label, string} @c -- @cindex \label\ @subentry \string\ @cindex \label\, \string\ @end macro @macro gcctabopt{body} @code{\body\} @end macro @c -- Get the version information --------------------------------------------- @include version.texi @c -- Entry for the Info dir structure ---------------------------------------- @ifnottex @dircategory Software development @direntry * gprofng: (gprofng). The next generation profiling tool for Linux @end direntry @end ifnottex @c -- Copyright stuff --------------------------------------------------------- @copying This document is the manual for @ProductName{}, last updated @value{UPDATED}. Copyright @copyright{} 2022-2023 Free Software Foundation, Inc. @c -- @quotation Permission is granted to copy, distribute and/or modify this document under the terms of the GNU Free Documentation License, Version 1.3 or any later version published by the Free Software Foundation; with no Invariant Sections, with no Front-Cover texts, and with no Back-Cover Texts. A copy of the license is included in the section entitled ``GNU Free Documentation License.'' @c -- @end quotation @end copying @finalout @smallbook @c -- Define the title page --------------------------------------------------- @titlepage @title GNU gprofng @subtitle The next generation profiling tool for Linux @subtitle version @value{VERSION} (last updated @value{UPDATED}) @author Ruud van der Pas @page @vskip 0pt plus 1filll @insertcopying @c man begin COPYRIGHT Copyright @copyright{} 2022-2023 Free Software Foundation, Inc. Permission is granted to copy, distribute and/or modify this document under the terms of the GNU Free Documentation License, Version 1.3 or any later version published by the Free Software Foundation; with no Invariant Sections, with no Front-Cover Texts, and with no Back-Cover Texts. A copy of the license is included in the section entitled ``GNU Free Documentation License''. @c man end @end titlepage @c -- Generate the Table of Contents ------------------------------------------ @contents @c -- The Top node ------------------------------------------------------------ @c Should contain a short summary, copying permissions and a master menu. @c ---------------------------------------------------------------------------- @ifnottex @node Top @top GNU Gprofng @insertcopying @end ifnottex @ifinfo @c -- The menu entries -------------------------------------------------------- @menu * Introduction:: About this manual. * Overview:: A brief overview of @ProductName{}. * A Mini Tutorial:: A short tutorial covering the key features. * Terminology:: Various concepts and some terminology explained. * Other Document Formats:: How to create this document in other formats. * Index:: The index. @detailmenu --- The Detailed Node Listing --- Introduction Overview * Main Features:: A high level overview. * Sampling versus Tracing:: The pros and cons of sampling versus tracing. * Steps Needed to Create a Profile:: How to create a profile. A Mini Tutorial * Getting Started:: The basics of profiling with @ProductName(). * Support for Multithreading:: Commands specific to multithreaded applications. * Viewing Multiple Experiments:: Analyze multiple experiments. * Profile Hardware Event Counters:: How to use hardware event counters. * Java Profiling:: How to profile a Java application. Terminology * The Program Counter:: What is a Program Counter? * Inclusive and Exclusive Metrics:: An explanation of inclusive and exclusive metrics. * Metric Definitions:: Definitions associated with metrics. * The Viewmode:: Select the way call stacks are presented. * The Selection List:: How to define a selection. * Load Objects and Functions:: The components in an application. * The Concept of a CPU in @ProductName{}:: The definition of a CPU. * Hardware Event Counters Explained:: What are event counters? * apath:: Our generic definition of a path. @c -- Index @end detailmenu @end menu @end ifinfo @ifset man @c man title gprofng the driver for the gprofng tool suite @c man begin SYNOPSIS gprofng [OPTION(S)] ACTION [@b{QUALIFIER}] [ARGUMENTS] TARGET @c man end @c man begin DESCRIPTION This is the driver for the GPROFNG tools suite to gather and analyze performance data. The driver executes the action specified. An example of an action is @code{collect} to collect performance data. Depending on the action, a qualifier may be needed to define the command. Several qualifiers support options. The last item on the command is the target the command applies to. For example, to collect performance data for an application called @code{a.out} and store the results in experiment directory @code{mydata.er}, the following command may be used: @smallexample $ gprofng collect app -o mydata.er a.out @end smallexample In this example, the action is @code{collect}, the qualifier is @code{app}, the single argument is @code{-o mydata.er} and the target is @code{a.out}. If gprofng is executed without any additional option, action, or target, a usage overview is printed. @c man end @c man begin OPTIONS @table @gcctabopt @item @var{--version} print the version number and exit. @item @var{--help} print usage information and exit. @end table @c man end @c man begin NOTES The gprofng driver supports the following commands. @c The man pages for the commands below can be viewed using the command name with "gprofng" replaced by "gp" and the spaces replaced by a dash ("-"). For example the man page @c name for "gprofng collect app" is "gp-collect-app". Collect performance data: @table @code @item gprofng collect app collect application performance data. @end table Display the performance results: @table @code @item gprofng display text display the performance data in ASCII format. @item gprofng display html generate an HTML file from one or more experiments. @end table Miscellaneous commands: @table @code @item gprofng display src display source or disassembly with compiler annotations. @item gprofng archive include binaries and source code in an experiment directory. @end table It is also possible to invoke the lower level commands directly, but since these are subject to change, in particular the options, we recommend to use the driver. @c man end @c man begin ENVIRONMENT The following environment variables are supported: @table @code @item @env{GPROFNG_MAX_CALL_STACK_DEPTH} set the depth of the call stack (default is 256). @item @env{GPROFNG_USE_JAVA_OPTIONS} may be set when profiling a C/C++ application that uses dlopen() to execute Java code. @item @env{GPROFNG_SSH_REMOTE_DISPLAY} use this variable to define the ssh command executed by the remote display tool. @item @env{GPROFNG_SKIP_VALIDATION} set this variable to disable checking hardware, system, and Java versions. @item @env{GPROFNG_ALLOW_CORE_DUMP} set this variable to allow a core file to be generated; otherwise an error report is created on /tmp. @item @env{GPROFNG_ARCHIVE} use this variable to define the settings for automatic archiving upon experiment recording completion. @item @env{GPROFNG_ARCHIVE_COMMON_DIR} set this variable to the location of the common archive. @item @env{GPROFNG_JAVA_MAX_CALL_STACK_DEPTH} set the depth of the Java call stack; the default is 256; set to 0 to disable capturing of call stacks. @item @env{GPROFNG_JAVA_NATIVE_MAX_CALL_STACK_DEPTH} set the depth of the Java native call stack; the default is 256; set to 0 to disable capturing of call stacks (JNI and assembly call stacks are not captured). @end table @c man end @c man begin SEEALSO The man pages for the various gprofng commands are not available yet, but the @option{--help} option supported on each of the commands lists the options and provides more information. For example this displays the options supported on the @command{gprofng collect app} command: @smallexample $ gprofng collect app --help @end smallexample The user guide is available as an Info entry for @file{gprofng}. @c man end @end ifset @c man begin DESCRIPTION @c man end @c -- A new node -------------------------------------------------------------- @node Introduction @chapter Introduction @c ---------------------------------------------------------------------------- The @ProductName{} tool is the next generation profiler for Linux. It consists of various commands to generate and display profile information. This manual starts with a tutorial how to create and interpret a profile. This part is highly practical and has the goal to get users up to speed as quickly as possible. As soon as possible, we would like to show you how to get your first profile on your screen. This is followed by more examples, covering many of the features. At the end of this tutorial, you should feel confident enough to tackle the more complex tasks. In a future update a more formal reference manual will be included as well. Since even in this tutorial we use certain terminology, we have included a chapter with descriptions at the end. In case you encounter unfamiliar wordings or terminology, please check this chapter. One word of caution. In several cases we had to somewhat tweak the screen output in order to make it fit. This is why the output may look somewhat different when you try things yourself. For now, we wish you a smooth profiling experience with @ProductName{} and good luck tackling performance bottlenecks. @c -- A new node -------------------------------------------------------------- @c cccccc @node A Brief Overview of @ProductName{} @node Overview @chapter A Brief Overview of @ProductName{} @c ---------------------------------------------------------------------------- @menu * Main Features:: A high level overview. * Sampling versus Tracing:: The pros and cons of sampling versus tracing. * Steps Needed to Create a Profile:: How to create a profile. @end menu Before we cover this tool in quite some detail, we start with a brief overview of what it is, and the main features. Since we know that many of you would like to get started rightaway, already in this first chapter we explain the basics of profiling with @ToolName{}. @c ---------------------------------------------------------------------------- @c TBD Review this text. Probably be more specific on the gcc releases and @c processor specifics. @c ---------------------------------------------------------------------------- @c -- A new node -------------------------------------------------------------- @node Main Features @section Main Features @c ---------------------------------------------------------------------------- @noindent These are the main features of the @ProductName{} tool: @itemize @bullet @item Profiling is supported for an application written in C, C++, Java, or Scala. @c TBD Java: up to 1.8 full support, support other than for modules @item Shared libraries are supported. The information is presented at the instruction level. @item The following multithreading programming models are supported: Pthreads, OpenMP, and Java threads. @item This tool works with unmodified production level executables. There is no need to recompile the code, but if the @code{-g} option has been used when building the application, source line level information is available. @item The focus is on support for code generated with the @code{gcc} compiler, but there is some limited support for the @code{icc} compiler as well. Future improvements and enhancements will focus on @code{gcc} though. @item Processors from Intel, AMD, and Arm are supported, but the level of support depends on the architectural details. In particular, hardware event counters may not be supported. @item Several views into the data are supported. For example, a function overview where the time is spent, but also a source line, disassembly, call tree and a caller-callees overview are available. @item Through filters, the user can zoom in on an area of interest. @item Two or more profiles can be aggregated, or used in a comparison. This comparison can be obtained at the function, source line, and disassembly level. @item Through a scripting language, and customization of the metrics shown, the generation and creation of a profile can be fully automated and provide tailored output. @end itemize @c -- A new node -------------------------------------------------------------- @node Sampling versus Tracing @section Sampling versus Tracing @c ---------------------------------------------------------------------------- A key difference with some other profiling tools is that the main data collection command @CollectApp{} mostly uses @cindex Program Counter sampling @cindex PC sampling Program Counter (PC) sampling under the hood. With @emph{sampling}, the executable is stopped at regular intervals. Each time it is halted, key information is gathered and stored. This includes the Program Counter that keeps track of where the execution is. Hence the name. Together with operational data, this information is stored in the experiment directory and can be viewed in the second phase. For example, the PC information is used to derive where the program was when it was halted. Since the sampling interval is known, it is relatively easy to derive how much time was spent in the various parts of the program. The opposite technique is generally referred to as @emph{tracing}. With tracing, the target is instrumented with specific calls that collect the requested information. These are some of the pros and cons of PC sampling verus tracing: @itemize @item Since there is no need to recompile, existing executables can be used and the profile measures the behaviour of exactly the same executable that is used in production runs. With sampling, one inherently profiles a different executable because the calls to the instrumentation library may affect the compiler optimizations and run time behaviour. @item With sampling, there are very few restrictions on what can be profiled and even without access to the source code, a basic profile can be made. @item A downside of sampling is that, depending on the sampling frequency, small functions may be missed or not captured accurately. Although this is rare, this may happen and is the reason why the user has control over the sampling rate. @item While tracing produces precise information, sampling is statistical in nature. As a result, small variations may occur across seemingly identical runs. We have not observed more than a few percent deviation though. Especially if the target job executed for a sufficiently long time. @item With sampling, it is not possible to get an accurate count how often functions are called. @end itemize @c -- A new node -------------------------------------------------------------- @node Steps Needed to Create a Profile @section Steps Needed to Create a Profile @c ---------------------------------------------------------------------------- Creating a profile takes two steps. First the profile data needs to be generated. This is followed by a viewing step to create a report from the information that has been gathered. Every @ProductName{} command starts with @ToolName{}, the name of the driver. This is followed by a keyword to define the high level functionality. Depending on this keyword, a third qualifier may be needed to further narrow down the request. This combination is then followed by options that are specific to the functionality desired. The command to gather, or ``collect'', the performance data is called @CollectApp{}. Aside from numerous options, this command takes the name of the target executable as an input parameter. Upon completion of the run, the performance data can be found in the newly created @cindex Experiment directory experiment directory. Unless explicitly specified otherwise, a default name for this directory is chosen. The name is @code{test..er} where @code{n} is the first integer number not in use yet for such a name. For example, the first time @CollectApp{} is invoked, an experiment directory with the name @code{test.1.er} is created. Upon a subsequent invocation of @CollectApp{} in the same directory, an experiment directory with the name @code{test.2.er} will be created, and so forth. Note that @CollectApp{} supports an option to explicitly name the experiment directory. Outside of the restriction that the name of this directory has to end with @code{.er}, any valid directory name can be used for this. Now that we have the performance data, the next step is to display it. @pindex @DisplayText{} The most commonly used command to view the performance information is @DisplayText{}. This is a very extensive and customizable tool that produces the information in ASCII format. @pindex @DisplayHTML{} Another option is to use @DisplayHTML{}. This tool generates a directory with files in html format. These can be viewed in a browser, allowing for easy navigation through the profile data. @c -- A new node -------------------------------------------------------------- @node A Mini Tutorial @chapter A Mini Tutorial @c ---------------------------------------------------------------------------- In this chapter we present and discuss the main functionality of @ToolName{}. This will be a practical approach, using an example code to generate profile data and show how to get various performance reports. @menu * Getting Started:: The basics of profiling with @ProductName(). * Support for Multithreading:: Commands specific to multithreaded applications. * Viewing Multiple Experiments:: Analyze multiple experiments. * Profile Hardware Event Counters:: How to use hardware event counters. * Java Profiling:: How to profile a Java application. @end menu @c -- A new node -------------------------------------------------------------- @node Getting Started @section Getting Started @c ---------------------------------------------------------------------------- The information presented here provides a good and common basis for many profiling tasks, but there are more features that you may want to leverage. These are covered in subsequent sections in this chapter. @menu * The Example Program:: A description of the example program used. * A First Profile:: How to get the first profile. * The Source Code View:: Display the metrics in the source code. * The Disassembly View:: Display the metrics at the instruction level. * Display and Define the Metrics:: An example how to customize the metrics. * A First Customization of the Output:: An example how to customize the output. * Name the Experiment Directory:: Change the name of the experiment directory. * Control the Number of Lines in the Output:: Change the number of lines in the tables. * Sorting the Performance Data:: How to set the metric to sort by. * Scripting:: Use a script to execute the commands. * A More Elaborate Example:: An example of customization. * The Call Tree:: Display the dynamic call tree. * More Information on the Experiment:: How to get additional statistics. * Control the Sampling Frequency:: How to control the sampling granularity. * Information on Load Objects:: How to get more information on load objects. @end menu @c -- A new node -------------------------------------------------------------- @node The Example Program @subsection The Example Program @c ---------------------------------------------------------------------------- Throughout this guide we use the same example C code that implements the multiplication of a vector of length @math{n} by an @math{m} by @math{n} matrix. The result is stored in a vector of length @math{m}. @cindex Pthreads @cindex Posix Threads The algorithm has been parallelized using Posix Threads, or Pthreads for short. The code was built using the @code{gcc} compiler and the name of the executable is @cindex mxv-pthreads.exe mxv-pthreads.exe. The matrix sizes can be set through the @code{-m} and @code{-n} options. The number of threads is set with the @code{-t} option. To increase the duration of the run, the multiplication is executed repeatedly. This is an example that multiplies a @math{3000} by @math{2000} matrix with a vector of length @math{2000} using @math{2} threads: @smallexample @verbatim $ ./mxv-pthreads.exe -m 3000 -n 2000 -t 2 mxv: error check passed - rows = 3000 columns = 2000 threads = 2 $ @end verbatim @end smallexample The program performs an internal check to verify the results are correct. The result of this check is printed, followed by the matrix sizes and the number of threads used. @c -- A new node -------------------------------------------------------------- @node A First Profile @subsection A First Profile @c ---------------------------------------------------------------------------- The first step is to collect the performance data. It is important to remember that much more information is gathered than may be shown by default. Often a single data collection run is sufficient to get a lot of insight. The @CollectApp{} command is used for the data collection. Nothing needs to be changed in the way the application is executed. The only difference is that it is now run under control of the tool, as shown below: @cartouche @smallexample $ gprofng collect app ./mxv.pthreads.exe -m 3000 -n 2000 -t 1 @end smallexample @end cartouche This command produces the following output: @smallexample @verbatim Creating experiment database test.1.er (Process ID: 2416504) ... mxv: error check passed - rows = 3000 columns = 2000 threads = 1 @end verbatim @end smallexample We see the message that a directory with the name @code{test.1.er} has been created. The application then completes as usual and we have our first experiment directory that can be analyzed. The tool we use for this is called @DisplayText{}. It takes the name of the experiment directory as an argument. @cindex Interpreter mode If invoked this way, the tool starts in the interactive @emph{interpreter} mode. While in this environment, commands can be given and the tool responds. This is illustrated below: @smallexample @verbatim $ gprofng display text test.1.er Warning: History and command editing is not supported on this system. (gp-display-text) quit $ @end verbatim @end smallexample @cindex Command line mode While useful in certain cases, we prefer to use this tool in command line mode, by specifying the commands to be issued when invoking the tool. The way to do this is to prepend the command with a hyphen (@code{-}) if used on the command line. For example, @IndexSubentry{Commands, @code{functions}} with the @code{functions} command we request a list of the functions that have been executed and their respective CPU times: @cartouche @smallexample $ gprofng display text -functions test.1.er @end smallexample @end cartouche @smallexample @verbatim $ gprofng display text -functions test.1.er Functions sorted by metric: Exclusive Total CPU Time Excl. Incl. Name Total Total CPU sec. CPU sec. 2.272 2.272 2.160 2.160 mxv_core 0.047 0.103 init_data 0.030 0.043 erand48_r 0.013 0.013 __drand48_iterate 0.013 0.056 drand48 0.008 0.010 _int_malloc 0.001 0.001 brk 0.001 0.002 sysmalloc 0. 0.001 __default_morecore 0. 0.113 __libc_start_main 0. 0.010 allocate_data 0. 2.160 collector_root 0. 2.160 driver_mxv 0. 0.113 main 0. 0.010 malloc 0. 0.001 sbrk @end verbatim @end smallexample As easy and simple as these steps are, we do have a first profile of our program! There are three columns. The first two contain the @cindex Total CPU time @emph{Total CPU Time}, which is the sum of the user and system time. @xref{Inclusive and Exclusive Metrics} for an explanation of ``exclusive'' and ``inclusive'' times. The first line echoes the metric that is used to sort the output. By default, this is the exclusive CPU time, but the sort metric can be changed by the user. We then see three columns with the exclusive and inclusive CPU times, plus the name of the function. @IndexSubentry{Miscellaneous, @code{}} The function with the name @code{} is not a user function, but is introduced by @ToolName{} and is used to display the accumulated metric values. In this case, we see that the total CPU time of this job was @code{2.272} seconds. With @code{2.160} seconds, function @code{mxv_core} is the most time consuming function. It is also a leaf function. The next function in the list is @code{init_data}. Although the CPU time spent in this part is negligible, this is an interesting entry because the inclusive CPU time of @code{0.103} seconds is higher than the exclusive CPU time of @code{0.047} seconds. Clearly it is calling another function, or even more than one function. @xref{The Call Tree} for the details how to get more information on this. The function @code{collector_root} does not look familiar. It is one of the internal functions used by @CollectApp{} and can be ignored. While the inclusive time is high, the exclusive time is zero. This means it doesn't contribute to the performance. The question is how we know where this function originates from? There is a very useful command to get more details on a function. @xref{Information on Load Objects}. @c -- A new node -------------------------------------------------------------- @node The Source Code View @subsection The Source Code View @c ---------------------------------------------------------------------------- In general, you would like to focus the tuning efforts on the most time consuming part(s) of the program. In this case that is easy, since 2.160 seconds on a total of 2.272 seconds is spent in function @code{mxv_core}. That is 95% of the total and it is time to dig deeper and look @cindex Source level timings at the time distribution at the source code level. @IndexSubentry{Commands, @code{source}} The @code{source} command is used to accomplish this. It takes the name of the function, not the source filename, as an argument. This is demonstrated below, where the @DisplayText{} command is used to show the annotated source listing of function @code{mxv_core}. Please note that the source code has to be compiled with the @code{-g} option in order for the source code feature to work. Otherwise the location can not be determined. @cartouche @smallexample $ gprofng display text -source mxv_core test.1.er @end smallexample @end cartouche The slightly modified output is as follows: @smallexample @verbatim Source file: /mxv.c Object file: mxv-pthreads.exe (found as test.1.er/archives/...) Load Object: mxv-pthreads.exe (found as test.1.er/archives/...) Excl. Incl. Total Total CPU sec. CPU sec. 0. 0. 32. void __attribute__ ((noinline)) mxv_core ( uint64_t row_index_start, uint64_t row_index_end, uint64_t m, uint64_t n, double **restrict A, double *restrict b, double *restrict c) 0. 0. 33. { 0. 0. 34. for (uint64_t i=row_index_start; i<=row_index_end; i++) { 0. 0. 35. double row_sum = 0.0; ## 1.687 1.687 36. for (int64_t j=0; j 1.687 1.687 mxv_core, line 36 in "mxv.c" 0.473 0.473 mxv_core, line 37 in "mxv.c" 0.032 0.088 init_data, line 72 in "manage_data.c" 0.030 0.043 0.013 0.013 0.013 0.056 0.012 0.012 init_data, line 77 in "manage_data.c" 0.008 0.010 0.003 0.003 init_data, line 71 in "manage_data.c" @end verbatim @end smallexample What this overview immediately highlights is that the next most time consuming source line takes 0.032 seconds only. With an inclusive time of 0.088 seconds, it is also clear that this branch of the code does not impact the performance. @c -- A new node -------------------------------------------------------------- @node The Disassembly View @subsection The Disassembly View @c ---------------------------------------------------------------------------- The source view is very useful to obtain more insight where the time is spent, but sometimes this is not sufficient. This is when the disassembly view comes in. It is activated with the @IndexSubentry{Commands, @code{disasm}} @code{disasm} command and as with the source view, it displays an annotated listing. In this @cindex Instruction level timings case it shows the instructions with the metrics, interleaved with the source lines. The instructions have a reference in square brackets (@code{[} and @code{]}) to the source line they correspond to. @noindent This is what we get for our example: @cartouche @smallexample $ gprofng display text -disasm mxv_core test.1.er @end smallexample @end cartouche @smallexample @verbatim Source file: /mxv.c Object file: mxv-pthreads.exe (found as test.1.er/archives/...) Load Object: mxv-pthreads.exe (found as test.1.er/archives/...) Excl. Incl. Total Total CPU sec. CPU sec. 32. void __attribute__ ((noinline)) mxv_core ( uint64_t row_index_start, uint64_t row_index_end, uint64_t m, uint64_t n, double **restrict A, double *restrict b, double *restrict c) 33. { 0. 0. [33] 4021ba: mov 0x8(%rsp),%r10 34. for (uint64_t i=row_index_start; i<=row_index_end; i++) { 0. 0. [34] 4021bf: cmp %rsi,%rdi 0. 0. [34] 4021c2: jbe 0x37 0. 0. [34] 4021c4: ret 35. double row_sum = 0.0; 36. for (int64_t j=0; j 1.683 1.683 mxv_core + 0x00000027, line 36 in "mxv.c" 0.375 0.375 mxv_core + 0x00000023, line 37 in "mxv.c" 0.096 0.096 mxv_core + 0x0000001D, line 37 in "mxv.c" 0.027 0.027 init_data + 0x000000BD, line 72 in "manage_data.c" 0.012 0.012 init_data + 0x00000117, line 77 in "manage_data.c" 0.008 0.008 _int_malloc + 0x00000A45 0.007 0.007 erand48_r + 0x00000062 0.006 0.006 drand48 + 0x00000000 0.005 0.005 __drand48_iterate + 0x00000005 @end verbatim @end smallexample @c -- A new node -------------------------------------------------------------- @node Display and Define the Metrics @subsection Display and Define the Metrics @c ---------------------------------------------------------------------------- The default metrics shown by @DisplayText{} are useful, but there is more recorded than displayed. We can customize the values shown by defining the metrics ourselves. @IndexSubentry{Commands, @code{metric_list}} There are two commands related to changing the metrics shown: @code{metric_list} and @IndexSubentry{Commands, @code{metrics}} @code{metrics}. The first command shows the metrics in use, plus all the metrics that have been stored as part of the experiment. The second command may be used to define the metric list. In our example we get the following values for the metrics: @IndexSubentry{Commands, @code{metric_list}} @cartouche @smallexample $ gprofng display text -metric_list test.1.er @end smallexample @end cartouche @smallexample @verbatim Current metrics: e.totalcpu:i.totalcpu:name Current Sort Metric: Exclusive Total CPU Time ( e.totalcpu ) Available metrics: Exclusive Total CPU Time: e.%totalcpu Inclusive Total CPU Time: i.%totalcpu Size: size PC Address: address Name: name @end verbatim @end smallexample This shows the metrics currently in use, the metric that is used to sort the data and all the metrics that have been recorded, but are not necessarily shown. @cindex Default metrics In this case, the default metrics are set to the exclusive and inclusive total CPU times, plus the name of the function, or load object. @IndexSubentry{Commands, @code{metrics}} The @code{metrics} command is used to define the metrics that need to be displayed. For example, to display the exclusive total CPU time, both as a number and a percentage, use the following metric definition: @code{e.%totalcpu} Since the metrics can be tailored for different views, there is a way to reset them to the default. This is done through the special keyword @code{default}. @c -- A new node -------------------------------------------------------------- @node A First Customization of the Output @subsection A First Customization of the Output @c ---------------------------------------------------------------------------- With the information just given, we can customize the function overview. For sake of the example, we would like to display the name of the function first, followed by the exclusive CPU time, given as an absolute number and a percentage. Note that the commands are parsed in order of appearance. This is why we need to define the metrics @emph{before} requesting the function overview: @cartouche @smallexample $ gprofng display text -metrics name:e.%totalcpu -functions test.1.er @end smallexample @end cartouche @smallexample @verbatim Current metrics: name:e.%totalcpu Current Sort Metric: Exclusive Total CPU Time ( e.%totalcpu ) Functions sorted by metric: Exclusive Total CPU Time Name Excl. Total CPU sec. % 2.272 100.00 mxv_core 2.160 95.04 init_data 0.047 2.06 erand48_r 0.030 1.32 __drand48_iterate 0.013 0.57 drand48 0.013 0.57 _int_malloc 0.008 0.35 brk 0.001 0.04 sysmalloc 0.001 0.04 __default_morecore 0. 0. __libc_start_main 0. 0. allocate_data 0. 0. collector_root 0. 0. driver_mxv 0. 0. main 0. 0. malloc 0. 0. sbrk 0. 0. @end verbatim @end smallexample This was a first and simple example how to customize the output. Note that we did not rerun our profiling job and merely modified the display settings. Below we will show other and also more advanced examples of customization. @c -- A new node -------------------------------------------------------------- @node Name the Experiment Directory @subsection Name the Experiment Directory @c ---------------------------------------------------------------------------- When using @CollectApp{}, the default names for experiments work fine, but they are quite generic. It is often more convenient to select a more descriptive name. For example, one that reflects conditions for the experiment conducted. For this, the mutually exclusive @code{-o} and @code{-O} options come in handy. Both may be used to provide a name for the experiment directory, but the behaviour of @CollectApp{} is different. With the @IndexSubentry{Options, @code{-o}} @code{-o} option, an existing experiment directory is not overwritten. You either need to explicitly remove an existing directory first, or use a name that is not in use yet. This is in contrast with the behaviour for the @IndexSubentry{Options, @code{-O}} @code{-O} option. Any existing (experiment) directory with the same name is silently overwritten. Be aware that the name of the experiment directory has to end with @code{.er}. @c -- A new node -------------------------------------------------------------- @node Control the Number of Lines in the Output @subsection Control the Number of Lines in the Output @c ---------------------------------------------------------------------------- @IndexSubentry{Commands, @code{limit}} The @code{limit } command can be used to control the number of lines printed in various overviews, including the function view, but it also takes effect for other display commands, like @code{lines}. The argument @code{} should be a positive integer number. It sets the number of lines in the function view. A value of zero resets the limit to the default. Be aware that the pseudo-function @code{} counts as a regular function. For example @code{limit 10} displays nine user level functions. @c -- A new node -------------------------------------------------------------- @node Sorting the Performance Data @subsection Sorting the Performance Data @c ---------------------------------------------------------------------------- @IndexSubentry{Commands, @code{sort}} The @code{sort } command sets the key to be used when sorting the performance data. The key is a valid metric definition, but the @cindex Visibility field visibility field (@xref{Metric Definitions}) in the metric definition is ignored since this does not affect the outcome of the sorting operation. For example if we set the sort key to @code{e.totalcpu}, the values will be sorted in descending order with respect to the exclusive total CPU time. The data can be sorted in reverse order by prepending the metric definition with a minus (@code{-}) sign. For example @code{sort -e.totalcpu}. A default metric for the sort operation has been defined and since this is a persistent command, this default can be restored with @code{default} as the key. @c -- A new node -------------------------------------------------------------- @node Scripting @subsection Scripting @c ---------------------------------------------------------------------------- As is probably clear by now, the list with commands for @DisplayText{} can be very long. This is tedious and also error prone. Luckily, there is an easier and more elegant way to control the behaviour of this tool. @IndexSubentry{Commands, @code{script}} Through the @code{script} command, the name of a file with commands can be passed in. These commands are parsed and executed as if they appeared on the command line in the same order as encountered in the file. The commands in this script file can actually be mixed with commands on the command line. The difference between the commands in the script file and those used on the command line is that the latter require a leading dash (@code{-}) symbol. Comment lines are supported. They need to start with the @code{#} symbol. @c -- A new node -------------------------------------------------------------- @node A More Elaborate Example @subsection A More Elaborate Example @c ---------------------------------------------------------------------------- With the information presented so far, we can customize our data gathering and display commands. As an example, to reflect the name of the algorithm and the number of threads that were used in the experiment, we select @code{mxv.1.thr.er} as the name of the experiment directory. All we then need to do is to add the @IndexSubentry{Options, @code{-O}} @code{-O} option followed by this name on the command line when running @CollectApp{}: @cartouche @smallexample $ exe=mxv-pthreads.exe $ m=3000 $ n=2000 $ gprofng collect app -O mxv.1.thr.er ./$exe -m $m -n $n -t 1 @end smallexample @end cartouche The commands to generate the profile are put into a file that we simply call @code{my-script}: @smallexample @verbatim $ cat my-script # This is my first gprofng script # Set the metrics metrics i.%totalcpu:e.%totalcpu:name # Use the exclusive time to sort sort e.totalcpu # Limit the function list to 5 lines limit 5 # Show the function list functions @end verbatim @end smallexample This script file is then specified as input to the @DisplayText{} command that is used to display the performance information stored in @code{mxv.1.thr.er}: @cartouche @smallexample $ gprofng display text -script my-script mxv.1.thr.er @end smallexample @end cartouche The command above produces the following output: @smallexample @verbatim # This is my first gprofng script # Set the metrics Current metrics: i.%totalcpu:e.%totalcpu:name Current Sort Metric: Exclusive Total CPU Time ( e.%totalcpu ) # Use the exclusive time to sort Current Sort Metric: Exclusive Total CPU Time ( e.%totalcpu ) # Limit the function list to 5 lines Print limit set to 5 # Show the function list Functions sorted by metric: Exclusive Total CPU Time Incl. Total Excl. Total Name CPU CPU sec. % sec. % 2.272 100.00 2.272 100.00 2.159 95.00 2.159 95.00 mxv_core 0.102 4.48 0.054 2.37 init_data 0.035 1.54 0.025 1.10 erand48_r 0.048 2.11 0.013 0.57 drand48 @end verbatim @end smallexample In the first part of the output, our comment lines in the script file are shown. These are interleaved with an acknowledgement message for the commands. This is followed by a profile consisting of 5 lines only. For both metrics, the percentages plus the timings are given. The numbers are sorted with respect to the exclusive total CPU time. It is now immediately clear that function @code{mxv_core} is responsbile for 95% of the CPU time and @code{init_data} takes 4.5% only. This is also where we see sampling in action. Although this is exactly the same job we profiled before, the timings are somewhat different, but the differences are very small. @c -- A new node -------------------------------------------------------------- @node The Call Tree @subsection The Call Tree @c ---------------------------------------------------------------------------- The call tree shows the dynamic hierarchy of the application by displaying the functions executed and their parent. It helps to find the most expensive path in the program. @IndexSubentry{Commands, @code{calltree}} This feature is enabled through the @code{calltree} command. This is how to get this tree for our current experiment: @cartouche @smallexample $ gprofng display text -calltree mxv.1.thr.er @end smallexample @end cartouche This displays the following structure: @smallexample @verbatim Functions Call Tree. Metric: Attributed Total CPU Time Attr. Name Total CPU sec. 2.272 +- 2.159 +-collector_root 2.159 | +-driver_mxv 2.159 | +-mxv_core 0.114 +-__libc_start_main 0.114 +-main 0.102 +-init_data 0.048 | +-drand48 0.035 | +-erand48_r 0.010 | +-__drand48_iterate 0.011 +-allocate_data 0.011 | +-malloc 0.011 | +-_int_malloc 0.001 | +-sysmalloc 0.001 +-check_results 0.001 +-malloc 0.001 +-_int_malloc @end verbatim @end smallexample At first sight this may not be what you expected and some explanation is in place. @c ---------------------------------------------------------------------------- @c TBD: Revise this text when we have user and machine mode. @c ---------------------------------------------------------------------------- First of all, function @code{collector_root} is internal to @ToolName{} and should be hidden to the user. This is part of a planned future enhancement. Recall that the @code{objects} and @code{fsingle} commands are very useful to find out more about load objects in general, but also to help identify an unknown entry in the function overview. @xref{Load Objects and Functions}. Another thing to note is that there are two main branches. The one under @code{collector_root} and the second one under @code{__libc_start_main}. This reflects the fact that we are executing a parallel program. Even though we only used one thread for this run, this is still executed in a separate path. The main, sequential part of the program is displayed under @code{main} and shows the functions called and the time they took. There are two things worth noting for the call tree feature: @itemize @item This is a dynamic tree and since sampling is used, it most likely looks slighlty different across seemingly identical profile runs. In case the run times are short, it is worth considering to use a high resolution through the @IndexSubentry{Options, @code{-p}} @code{-p} option. For example to use @code{-p hi} to increase the sampling rate. @item In case hardware event counters have been enabled (@xref{Profile Hardware Event Counters}), these values are also displayed in the call tree view. @end itemize @c -- A new node -------------------------------------------------------------- @node More Information on the Experiment @subsection More Information on the Experiment @c ---------------------------------------------------------------------------- The experiment directory not only contains performance related data. Several system characteristics, the actually command executed, and some global performance statistics can be displayed. @IndexSubentry{Commands, @code{header}} The @code{header} command displays information about the experiment(s). For example, this is the command to extract this data from for our experiment directory: @cartouche @smallexample $ gprofng display text -header mxv.1.thr.er @end smallexample @end cartouche The above command prints the following information. Note that some of the lay-out and the information has been modified. The textual changes are marked with the @code{<} and @code{>} symbols. @smallexample @verbatim Experiment: mxv.1.thr.er No errors No warnings Archive command `gp-archive -n -a on --outfile /archive.log ' Target command (64-bit): './mxv-pthreads.exe -m 3000 -n 2000 -t 1' Process pid 30591, ppid 30589, pgrp 30551, sid 30468 Current working directory: Collector version: `2.36.50'; experiment version 12.4 (64-bit) Host `', OS `Linux ', page size 4096, architecture `x86_64' 16 CPUs, clock speed 1995 MHz. Memory: 30871514 pages @ 4096 = 120591 MB. Data collection parameters: Clock-profiling, interval = 997 microsecs. Periodic sampling, 1 secs. Follow descendant processes from: fork|exec|combo Experiment started Experiment Ended: 2.293162658 Data Collection Duration: 2.293162658 @end verbatim @end smallexample The output above may assist in troubleshooting, or to verify some of the operational conditions and we recommand to include this command when generating a profile. @IndexSubentry{Options, @code{-C}} Related to this command there is a useful option to record your own comment(s) in an experiment. To this end, use the @code{-C} option on the @CollectApp{} tool to specify a comment string. Up to ten comment lines can be included. These comments are displayed with the @code{header} command on the @DisplayText{} tool. @IndexSubentry{Commands, @code{overview}} The @code{overview} command displays information on the experiment(s) and also shows a summary of the values for the metric(s) used. This is an example how to use it on our newly created experiment directory: @cartouche @smallexample $ gprofng display text -overview mxv.1.thr.er @end smallexample @end cartouche @smallexample @verbatim Experiment(s): Experiment :mxv.1.thr.er Target : './mxv-pthreads.exe -m 3000 -n 2000 -t 1' Host : (, Linux ) Start Time : Duration : 2.293 Seconds Metrics: Experiment Duration (Seconds): [2.293] Clock Profiling [X]Total CPU Time - totalcpu (Seconds): [*2.272] Notes: '*' indicates hot metrics, '[X]' indicates currently enabled metrics. The metrics command can be used to change selections. The metric_list command lists all available metrics. @end verbatim @end smallexample This command provides a dashboard overview that helps to easily identify where the time is spent and in case hardware event counters are used, it shows their total values. @c -- A new node -------------------------------------------------------------- @node Control the Sampling Frequency @subsection Control the Sampling Frequency @c ---------------------------------------------------------------------------- So far we did not talk about the frequency of the sampling process, but in some cases it is useful to change the default of 10 milliseconds. The advantage of increasing the sampling frequency is that functions that do not take much time per invocation are more accurately captured. The downside is that more data is gathered. This has an impact on the overhead of the collection process and more disk space is required. In general this is not an immediate concern, but with heavily threaded applications that run for an extended period of time, increasing the frequency may have a more noticeable impact. @IndexSubentry{Options, @code{-p}} The @code{-p} option on the @CollectApp{} tool is used to enable or disable clock based profiling, or to explicitly set the sampling rate. @cindex Sampling interval This option takes one of the following keywords: @table @code @item off Disable clock based profiling. @item on Enable clock based profiling with a per thread sampling interval of 10 ms. This is the default. @item lo Enable clock based profiling with a per thread sampling interval of 100 ms. @item hi Enable clock based profiling with a per thread sampling interval of 1 ms. @item Enable clock based profiling with a per thread sampling interval of . @end table One may wonder why there is an option to disable clock based profiling. This is because by default, it is enabled when conducting hardware event counter experiments (@xref{Profile Hardware Event Counters}). With the @code{-p off} option, this can be disabled. If an explicit value is set for the sampling, the number can be an integer or a floating-point number. A suffix of @code{u} for microseconds, or @code{m} for milliseconds is supported. If no suffix is used, the value is assumed to be in milliseconds. If the value is smaller than the clock profiling minimum, a warning message is issued and it is set to the minimum. In case it is not a multiple of the clock profiling resolution, it is silently rounded down to the nearest multiple of the clock resolution. If the value exceeds the clock profiling maximum, is negative, or zero, an error is reported. @IndexSubentry{Commands, @code{header}} Note that the @code{header} command echoes the sampling rate used. @c -- A new node -------------------------------------------------------------- @node Information on Load Objects @subsection Information on Load Objects @c ---------------------------------------------------------------------------- It may happen that the function list contains a function that is not known to the user. This can easily happen with library functions for example. Luckily there are three commands that come in handy then. @IndexSubentry{Commands, @code{objects}} @IndexSubentry{Commands, @code{fsingle}} @IndexSubentry{Commands, @code{fsummary}} These commands are @code{objects}, @code{fsingle}, and @code{fsummary}. They provide details on @cindex Load objects load objects (@xref{Load Objects and Functions}). The @code{objects} command lists all load objects that have been referenced during the performance experiment. Below we show the command and the result for our profile job. Like before, the (long) path names in the output have been shortened and replaced by the @IndexSubentry{Miscellaneous, @code{}} @code{} symbol that represents an absolute directory path. @cartouche @smallexample $ gprofng display text -objects mxv.1.thr.er @end smallexample @end cartouche The output includes the name and path of the target executable: @smallexample @verbatim () (/mxv-pthreads.exe) (/usr/lib64/librt-2.17.so) (/usr/lib64/libdl-2.17.so) (/libbfd-2.36.50 ) (/libopcodes-2. ) (/usr/lib64/libc-2.17.so) (/usr/lib64/libpthread-2.17.so) (/usr/lib64/libm-2.17.so) (/libgp-collector.so) (/usr/lib64/ld-2.17.so) (DYNAMIC_FUNCTIONS) @end verbatim @end smallexample @IndexSubentry{Commands, @code{fsingle}} The @code{fsingle} command may be used to get more details on a specific entry in the function view, say. For example, the command below provides additional information on the @code{collector_root} function shown in the function overview. @cartouche @smallexample $ gprofng display text -fsingle collector_root mxv.1.thr.er @end smallexample @end cartouche Below the output from this command. It has been somewhat modified to match the display requirements. @smallexample @verbatim collector_root Exclusive Total CPU Time: 0. ( 0. %) Inclusive Total CPU Time: 2.159 ( 95.0%) Size: 401 PC Address: 10:0x0001db60 Source File: /dispatcher.c Object File: mxv.1.thr.er/archives/libgp-collector.so_HpzZ6wMR-3b Load Object: /libgp-collector.so Mangled Name: Aliases: @end verbatim @end smallexample In this table we not only see how much time was spent in this function, we also see where it originates from. In addition to this, the size and start address are given as well. If the source code location is known it is also shown here. @IndexSubentry{Commands, @code{fsummary}} The related @code{fsummary} command displays the same information as @code{fsingle}, but for all functions in the function overview, including @code{}: @cartouche @smallexample $ gprofng display text -fsummary mxv.1.thr.er @end smallexample @end cartouche @smallexample @verbatim Functions sorted by metric: Exclusive Total CPU Time Exclusive Total CPU Time: 2.272 (100.0%) Inclusive Total CPU Time: 2.272 (100.0%) Size: 0 PC Address: 1:0x00000000 Source File: (unknown) Object File: (unknown) Load Object: Mangled Name: Aliases: mxv_core Exclusive Total CPU Time: 2.159 ( 95.0%) Inclusive Total CPU Time: 2.159 ( 95.0%) Size: 75 PC Address: 2:0x000021ba Source File: /mxv.c Object File: mxv.1.thr.er/archives/mxv-pthreads.exe_hRxWdccbJPc Load Object: /mxv-pthreads.exe Mangled Name: Aliases: ... etc ... @end verbatim @end smallexample @c -- A new node -------------------------------------------------------------- @node Support for Multithreading @section Support for Multithreading @c ---------------------------------------------------------------------------- In this chapter we introduce and discuss the support for multithreading. As is shown below, nothing needs to be changed when collecting the performance data. The difference is that additional commands are available to get more information on the parallel environment, plus that several filters allow the user to zoom in on specific threads. @c -- A new node -------------------------------------------------------------- @node Creating a Multithreading Experiment @subsection Creating a Multithreading Experiment @c ---------------------------------------------------------------------------- We demonstrate the support for multithreading using the same code and settings as before, but this time we use 2 threads: @cartouche @smallexample $ exe=mxv-pthreads.exe $ m=3000 $ n=2000 $ gprofng collect app -O mxv.2.thr.er ./$exe -m $m -n $n -t 2 @end smallexample @end cartouche First of all, note that we did not change anything, other than setting the number of threads to 2. Nothing special is needed to profile a multithreaded job when using @ToolName{}. The same is true when displaying the performance results. The same commands that we used before work unmodified. For example, this is all that is needed to get a function overview: @cartouche @smallexample $ gpprofng display text -limit 10 -functions mxv.2.thr.er @end smallexample @end cartouche This produces the following familiar looking output: @smallexample @verbatim Print limit set to 10 Functions sorted by metric: Exclusive Total CPU Time Excl. Incl. Name Total Total CPU sec. CPU sec. 2.268 2.268 2.155 2.155 mxv_core 0.044 0.103 init_data 0.030 0.046 erand48_r 0.016 0.016 __drand48_iterate 0.013 0.059 drand48 0.008 0.011 _int_malloc 0.003 0.003 brk 0. 0.003 __default_morecore 0. 0.114 __libc_start_main @end verbatim @end smallexample @c -- A new node -------------------------------------------------------------- @node Commands Specific to Multithreading @subsection Commands Specific to Multithreading @c ---------------------------------------------------------------------------- The function overview shown above shows the results aggregated over all the threads. The interesting new element is that we can also look at the performance data for the individual threads. @IndexSubentry{Commands, @code{thread_list}} The @code{thread_list} command displays how many threads have been used: @cartouche @smallexample $ gprofng display text -thread_list mxv.2.thr.er @end smallexample @end cartouche This produces the following output, showing that three threads have been used: @smallexample @verbatim Exp Sel Total === === ===== 1 all 3 @end verbatim @end smallexample The output confirms there is one experiment and that by default all threads are selected. It may seem surprising to see three threads here, since we used the @code{-t 2} option, but it is common for a Pthreads program to use one additional thread. This is typically the thread that runs from start to finish and handles the sequential portions of the code, as well as takes care of managing the threads. It is no different in our example code. At some point, the main thread creates and activates the two threads that perform the multiplication of the matrix with the vector. Upon completion of this computation, the main thread continues. @IndexSubentry{Commands, @code{threads}} The @code{threads} command is simple, yet very powerful. It shows the total value of the metrics for each thread. To make it easier to interpret the data, we modify the metrics to include percentages: @cartouche @smallexample $ gprofng display text -metrics e.%totalcpu -threads mxv.2.thr.er @end smallexample @end cartouche The command above produces the following overview: @smallexample @verbatim Current metrics: e.%totalcpu:name Current Sort Metric: Exclusive Total CPU Time ( e.%totalcpu ) Objects sorted by metric: Exclusive Total CPU Time Excl. Total Name CPU sec. % 2.258 100.00 1.075 47.59 Process 1, Thread 3 1.070 47.37 Process 1, Thread 2 0.114 5.03 Process 1, Thread 1 @end verbatim @end smallexample The first line gives the total CPU time accumulated over the threads selected. This is followed by the metric value(s) for each thread. From this it is clear that the main thread is responsible for 5% of the total CPU time, while the other two threads take 47% each. This view is ideally suited to verify if there any load balancing issues and also to find the most time consuming thread(s). @IndexSubentry{Filters, Thread selection} While useful, often more information than this is needed. This is @IndexSubentry{Commands, @code{thread_select}} where the thread selection filter comes in. Through the @code{thread_select} command, one or more threads may be selected (@xref{The Selection List} how to define the selection list). Since it is most common to use this command in a script, we do so as well here. Below the script we are using: @cartouche @smallexample # Define the metrics metrics e.%totalcpu # Limit the output to 10 lines limit 10 # Get the function overview for thread 1 thread_select 1 functions # Get the function overview for thread 2 thread_select 2 functions # Get the function overview for thread 3 thread_select 3 functions @end smallexample @end cartouche The definition of the metrics and the output limiter has been shown and explained before and will be ignored. The new command we focus on is @IndexSubentry{Commands, @code{thread_select}} @code{thread_select}. This command takes a list (@xref{The Selection List}) to select specific threads. In this case we simply use the individual thread numbers that we obtained with the @code{thread_list} command earlier. This restricts the output of the @code{functions} command to the thread number(s) specified. This means that the script above shows which function(s) each thread executes and how much CPU time they consumed. Both the timings and their percentages are given. This is the relevant part of the output for the first thread: @smallexample @verbatim # Get the function overview for thread 1 Exp Sel Total === === ===== 1 1 3 Functions sorted by metric: Exclusive Total CPU Time Excl. Total Name CPU sec. % 0.114 100.00 0.051 44.74 init_data 0.028 24.56 erand48_r 0.017 14.91 __drand48_iterate 0.010 8.77 _int_malloc 0.008 7.02 drand48 0. 0. __libc_start_main 0. 0. allocate_data 0. 0. main 0. 0. malloc @end verbatim @end smallexample As usual, the comment lines are echoed. This is followed by a confirmation of our selection. We see that indeed thread 1 has been selected. What is displayed next is the function overview for this particular thread. Due to the @code{limit 10} command, there are ten entries in this list. Below are the overviews for threads 2 and 3 respectively. We see that all of the CPU time is spent in function @code{mxv_core} and that this time is approximately the same for both threads. @smallexample @verbatim # Get the function overview for thread 2 Exp Sel Total === === ===== 1 2 3 Functions sorted by metric: Exclusive Total CPU Time Excl. Total Name CPU sec. % 1.072 100.00 1.072 100.00 mxv_core 0. 0. collector_root 0. 0. driver_mxv # Get the function overview for thread 3 Exp Sel Total === === ===== 1 3 3 Functions sorted by metric: Exclusive Total CPU Time Excl. Total Name CPU sec. % 1.076 100.00 1.076 100.00 mxv_core 0. 0. collector_root 0. 0. driver_mxv @end verbatim @end smallexample When analyzing the performance of a multithreaded application, it is sometimes useful to know whether threads have mostly executed on the same core, say, or if they have wandered across multiple cores. This sort of stickiness is usually referred to as @cindex Thread affinity @emph{thread affinity}. Similar to the commands for the threads, there are several commands related to the usage of the cores, or @emph{CPUs} as they are called in @ToolName{} (@xref{The Concept of a CPU in @ProductName{}}). In order to have some more interesting data to look at, we created a new experiment, this time using 8 threads: @cartouche @smallexample $ exe=mxv-pthreads.exe $ m=3000 $ n=2000 $ gprofng collect app -O mxv.8.thr.er ./$exe -m $m -n $n -t 8 @end smallexample @end cartouche @IndexSubentry{Commands, @code{cpu_list}} Similar to the @code{thread_list} command, the @code{cpu_list} command displays how many CPUs have been used. @IndexSubentry{Commands, @code{cpus}} The equivalent of the @code{threads} threads command, is the @code{cpus} command, which shows the CPU numbers that were used and how much time was spent on each of them. Both are demonstrated below. @cartouche @smallexample $ gprofng display text -metrics e.%totalcpu -cpu_list -cpus mxv.8.thr.er @end smallexample @end cartouche This command produces the following output: @smallexample @verbatim Current metrics: e.%totalcpu:name Current Sort Metric: Exclusive Total CPU Time ( e.%totalcpu ) Exp Sel Total === === ===== 1 all 10 Objects sorted by metric: Exclusive Total CPU Time Excl. Total Name CPU sec. % 2.310 100.00 0.286 12.39 CPU 7 0.284 12.30 CPU 13 0.282 12.21 CPU 5 0.280 12.13 CPU 14 0.266 11.52 CPU 9 0.265 11.48 CPU 2 0.264 11.44 CPU 11 0.194 8.42 CPU 0 0.114 4.92 CPU 1 0.074 3.19 CPU 15 @end verbatim @end smallexample @c ---------------------------------------------------------------------------- @c TBD - Ruud @c I'd like to improve this and have a way to see where a thread has executed. @c ---------------------------------------------------------------------------- What we see in this table is that a total of 10 CPUs have been used. This is followed by a list with all the CPU numbers that have been used during the run. For each CPU it is shown how much time was spent on it. While the table with thread times shown earlier may point at a load imbalance in the application, this overview has a different purpose. For example, we see that 10 CPUs have been used, but we know that the application uses 9 threads only. This means that at least one thread has executed on more than one CPU. In itself this is not something to worry about, but warrants a deeper investigation. Honesty dictates that next we performed a pre-analysis to find out which thread(s) have been running on more than one CPU. We found this to be thread 7. It has executed on CPUs 0 and 15. With this knowledge, we wrote the script shown below. It zooms in on the behaviour of thread 7. @cartouche @smallexample # Define the metrics metrics e.%totalcpu # Limit the output to 10 lines limit 10 functions # Get the function overview for CPU 0 cpu_select 0 functions # Get the function overview for CPU 15 cpu_select 15 functions @end smallexample @end cartouche From the earlier shown threads overview, we know that thread 7 has used @code{0.268} seconds of CPU time.. By selecting CPUs 0 and 15, respectively, we get the following function overviews: @smallexample @verbatim # Get the function overview for CPU 0 Exp Sel Total === === ===== 1 0 10 Functions sorted by metric: Exclusive Total CPU Time Excl. Total Name CPU sec. % 0.194 100.00 0.194 100.00 mxv_core 0. 0. collector_root 0. 0. driver_mxv # Get the function overview for CPU 15 Exp Sel Total === === ===== 1 15 10 Functions sorted by metric: Exclusive Total CPU Time Excl. Total Name CPU sec. % 0.074 100.00 0.074 100.00 mxv_core 0. 0. collector_root 0. 0. driver_mxv @end verbatim @end smallexample This shows that thread 7 spent @code{0.194} seconds on CPU 0 and @code{0.074} seconds on CPU 15. @c -- A new node -------------------------------------------------------------- @node Viewing Multiple Experiments @section Viewing Multiple Experiments @c ---------------------------------------------------------------------------- One thing we did not cover sofar is that @ToolName{} fully supports the analysis of multiple experiments. The @DisplayText{} tool accepts a list of experiments. The data can either be aggregated across the experiments, or used in a comparison. Mention @code{experiment_list} @c -- A new node -------------------------------------------------------------- @node Aggregation of Experiments @subsection Aggregation of Experiments @c ---------------------------------------------------------------------------- By default, the data for multiple experiments is aggregrated and the display commands shows these combined results. For example, we can aggregate the data for our single and dual thread experiments. Below is the script we used for this: @cartouche @smallexample # Define the metrics metrics e.%totalcpu # Limit the output to 10 lines limit 10 # Get the list with experiments experiment_list # Get the function overview functions @end smallexample @end cartouche @IndexSubentry{Commands, @code{experiment_list}} With the exception of the @code{experiment_list} command, all commands used have been discussed earlier. The @code{experiment_list} command provides a list of the experiments that have been loaded. This is is used to verify we are looking at the experiments we intend to aggregate. @cartouche @smallexample $ gprofng display text -script my-script-agg mxv.1.thr.er mxv.2.thr.er @end smallexample @end cartouche With the command above, we get the following output: @smallexample @verbatim # Define the metrics Current metrics: e.%totalcpu:name Current Sort Metric: Exclusive Total CPU Time ( e.%totalcpu ) # Limit the output to 10 lines Print limit set to 10 # Get the list with experiments ID Sel PID Experiment == === ===== ============ 1 yes 30591 mxv.1.thr.er 2 yes 11629 mxv.2.thr.er # Get the function overview Functions sorted by metric: Exclusive Total CPU Time Excl. Total Name CPU sec. % 4.533 100.00 4.306 94.99 mxv_core 0.105 2.31 init_data 0.053 1.17 erand48_r 0.027 0.59 __drand48_iterate 0.021 0.46 _int_malloc 0.021 0.46 drand48 0.001 0.02 sysmalloc 0. 0. __libc_start_main 0. 0. allocate_data @end verbatim @end smallexample The first five lines should look familiar. The five lines following, echo the comment line in the script and show the overview of the experiments. This confirms two experiments have been loaded and that both are active. This is followed by the function overview. The timings have been summed up and the percentages are adjusted accordingly. For example, the total accumulated time is indeed 2.272 + 2.261 = 4.533 seconds. @c -- A new node -------------------------------------------------------------- @node Comparison of Experiments @subsection Comparison of Experiments @c ---------------------------------------------------------------------------- The support for multiple experiments really shines in comparison mode. This feature is enabled through the command @IndexSubentry{Commands, @code{compare on/off}} @code{compare on} and is disabled by setting @code{compare off}. @cindex Compare experiments In comparison mode, the data for the various experiments is shown side by side, as illustrated below where we compare the results for the multithreaded experiments using one and two threads respectively: @cartouche @smallexample $ gprofng display text -compare on -functions mxv.1.thr.er mxv.2.thr.er @end smallexample @end cartouche @noindent This produces the following output: @smallexample @verbatim Functions sorted by metric: Exclusive Total CPU Time mxv.1.thr.er mxv.2.thr.er mxv.1.thr.er mxv.2.thr.er Excl. Total Excl. Total Incl. Total Incl. Total Name CPU CPU CPU CPU sec. sec. sec. sec. 2.272 2.261 2.272 2.261 2.159 2.148 2.159 2.148 mxv_core 0.054 0.051 0.102 0.104 init_data 0.025 0.028 0.035 0.045 erand48_r 0.013 0.008 0.048 0.053 drand48 0.011 0.010 0.012 0.010 _int_malloc 0.010 0.017 0.010 0.017 __drand48_iterate 0.001 0. 0.001 0. sysmalloc 0. 0. 0.114 0.114 __libc_start_main 0. 0. 0.011 0.010 allocate_data 0. 0. 0.001 0. check_results 0. 0. 2.159 2.148 collector_root 0. 0. 2.159 2.148 driver_mxv 0. 0. 0.114 0.114 main 0. 0. 0.012 0.010 malloc @end verbatim @end smallexample This table is already helpful to more easily compare (two) profiles, but there is more that we can do here. By default, in comparison mode, all measured values are shown. Often profiling is about comparing performance data. It is therefore more useful to look at differences, or ratios, using one experiment as a reference. The values shown are relative to this difference. For example if a ratio is below one, it means the reference value was higher. @IndexSubentry{Commands, @code{compare on/off}} This feature is supported on the @code{compare} command. In addition to @code{on}, or @code{off}, this command also supports @IndexSubentry{Commands, @code{compare delta}} @code{delta}, or @IndexSubentry{Commands, @code{compare ratio}} @code{ratio}. Usage of one of these two keywords enables the comparison feature and shows either the difference, or the ratio, relative to the reference data. In the example below, we use the same two experiments used in the comparison above, but as before, the number of lines is restricted to 10 and we focus on the exclusive timings plus percentages. For the comparison part we are interested in the differences. This is the script that produces such an overview: @cartouche @smallexample # Define the metrics metrics e.%totalcpu # Limit the output to 10 lines limit 10 # Set the comparison mode to differences compare delta # Get the function overview functions @end smallexample @end cartouche Assuming this script file is called @code{my-script-comp}, this is how we get the table displayed on our screen: @cartouche @smallexample $ gprofng display text -script my-script-comp mxv.1.thr.er mxv.2.thr.er @end smallexample @end cartouche Leaving out some of the lines printed, but we have seen before, we get the following table: @smallexample @verbatim mxv.1.thr.er mxv.2.thr.er Excl. Total Excl. Total Name CPU CPU sec. % delta % 2.272 100.00 -0.011 100.00 2.159 95.00 -0.011 94.97 mxv_core 0.054 2.37 -0.003 2.25 init_data 0.025 1.10 +0.003 1.23 erand48_r 0.013 0.57 -0.005 0.35 drand48 0.011 0.48 -0.001 0.44 _int_malloc 0.010 0.44 +0.007 0.75 __drand48_iterate 0.001 0.04 -0.001 0. sysmalloc 0. 0. +0. 0. __libc_start_main 0. 0. +0. 0. allocate_data @end verbatim @end smallexample It is now easy to see that the CPU times for the most time consuming functions in this code are practically the same. While in this case we used the delta as a comparison, Note that the comparison feature is supported at the function, source, and disassembly level. There is no practical limit on the number of experiments that can be used in a comparison. @c -- A new node -------------------------------------------------------------- @node Profile Hardware Event Counters @section Profile Hardware Event Counters @c ---------------------------------------------------------------------------- Many processors provide a set of hardware event counters and @ToolName{} provides support for this feature. @xref{Hardware Event Counters Explained} for those readers that are not familiar with such counters and like to learn more. In this section we explain how to get the details on the event counter support for the processor used in the experiment(s), and show several examples. @c -- A new node -------------------------------------------------------------- @node Getting Information on the Counters Supported @subsection Getting Information on the Counters Supported @c ---------------------------------------------------------------------------- The first step is to check if the processor used for the experiments is supported by @ToolName{}. @IndexSubentry{Options, @code{-h}} The @code{-h} option on @CollectApp{} will show the event counter information: @cartouche @smallexample $ gprofng collect app -h @end smallexample @end cartouche In case the counters are supported, a list with the events is printed. Otherwise, a warning message will be issued. For example, below we show this command and the output on an Intel Xeon Platinum 8167M (aka ``Skylake'') processor. The output has been split into several sections and each section is commented upon separately. @smallexample @verbatim Run "gprofng collect app --help" for a usage message. Specifying HW counters on `Intel Arch PerfMon v2 on Family 6 Model 85' (cpuver=2499): -h {auto|lo|on|hi} turn on default set of HW counters at the specified rate -h [-h ]... -h [,]... specify HW counter profiling for up to 4 HW counters @end verbatim @end smallexample The first line shows how to get a usage overview. This is followed by some information on the target processor. The next five lines explain in what ways the @code{-h} option can be used to define the events to be monitored. The first version shown above enables a default set of counters. This default depends on the processor this command is executed on. The keyword following the @code{-h} option defines the sampling rate: @table @code @item auto Match the sample rate of used by clock profiling. If the latter is disabled, Use a per thread sampling rate of approximately 100 samples per second. This setting is the default and preferred. @item on Use a per thread sampling rate of approximately 100 samples per second. @item lo Use a per thread sampling rate of approximately 10 samples per second. @item hi Use a per thread sampling rate of approximately 1000 samples per second. @end table The second and third variant define the events to be monitored. Note that the number of simultaneous events supported is printed. In this case we can monitor four events in a single profiling job. It is a matter of preference whether you like to use the @code{-h} option for each event, or use it once, followed by a comma separated list. There is one slight catch though. The counter definition below has mandatory comma (@code{,}) between the event and the rate. While a default can be used for the rate, the comma cannot be omitted. This may result in a somewhat awkward counter definition in case the default sampling rate is used. For example, the following two commands are equivalent. Note the double comma in the second command. This is not a typo. @cartouche @smallexample $ gprofng collect app -h cycles -h insts ... $ gprofng collect app -h cycles,,insts ... @end smallexample @end cartouche In the first command this comma is not needed, because a comma (``@code{,}'') immediately followed by white space may be omitted. This is why we prefer the this syntax and in the remainder will use the first version of this command. @IndexSubentry{Hardware event counters, counter definition} The counter definition takes an event name, plus optionally one or more attributes, followed by a comma, and optionally the sampling rate. The output section below shows the formal definition. @cartouche @smallexample == [[~=]...],[] @end smallexample @end cartouche The printed help then explains this syntax. Below we have summarized and expanded this output: @table @code @item The counter name must be selected from the available counters listed as part of the output printed with the @code{-h} option. On most systems, if a counter is not listed, it may still be specified by its numeric value. @item ~= This is an optional attribute that depends on the processor. The list of supported attributes is printed in the output. Examples of attributes are ``user'', or ``system''. The value can given in decimal or hexadecimal format. Multiple attributes may be specified, and each must be preceded by a ~. @item The sampling rate is one of the following: @table @code @item auto This is the default and matches the rate used by clock profiling. If clock profiling is disabled, use @code{on}. @item on Set the per thread maximum sampling rate to ~100 samples/second @item lo Set the per thread maximum sampling rate to ~10 samples/second @item hi Set the per thread maximum sampling rate to ~1000 samples/second @item Define the sampling interval. @xref{Control the Sampling Frequency} how to define this. @end table @end table After the section with the formal definition of events and counters, a processor specific list is displayed. This part starts with an overview of the default set of counters and the aliased names supported @emph{on this specific processor}. @smallexample @verbatim Default set of HW counters: -h cycles,,insts,,llm Aliases for most useful HW counters: alias raw name type units regs description cycles unhalted-core-cycles CPU-cycles 0123 CPU Cycles insts instruction-retired events 0123 Instructions Executed llm llc-misses events 0123 Last-Level Cache Misses br_msp branch-misses-retired events 0123 Branch Mispredict br_ins branch-instruction-retired events 0123 Branch Instructions @end verbatim @end smallexample The definitions given above may or may not be available on other processors, but we try to maximize the overlap across alias sets. The table above shows the default set of counters defined for this processor, and the aliases. For each alias the full ``raw'' name is given, plus the unit of the number returned by the counter (CPU cycles, or a raw count), the hardware counter the event is allowed to be mapped onto, and a short description. The last part of the output contains all the events that can be monitored: @smallexample @verbatim Raw HW counters: name type units regs description unhalted-core-cycles CPU-cycles 0123 unhalted-reference-cycles events 0123 instruction-retired events 0123 llc-reference events 0123 llc-misses events 0123 branch-instruction-retired events 0123 branch-misses-retired events 0123 ld_blocks.store_forward events 0123 ld_blocks.no_sr events 0123 ld_blocks_partial.address_alias events 0123 dtlb_load_misses.miss_causes_a_walk events 0123 dtlb_load_misses.walk_completed_4k events 0123 l2_lines_out.silent events 0123 l2_lines_out.non_silent events 0123 l2_lines_out.useless_hwpf events 0123 sq_misc.split_lock events 0123 See Chapter 19 of the "Intel 64 and IA-32 Architectures Software Developer's Manual Volume 3B: System Programming Guide" @end verbatim @end smallexample As can be seen, these names are not always easy to correlate to a specific event of interest. The processor manual should provide more clarity on this. @c -- A new node -------------------------------------------------------------- @node Examples Using Hardware Event Counters @subsection Examples Using Hardware Event Counters @c ---------------------------------------------------------------------------- The previous section may give the impression that these counters are hard to use, but as we will show now, in practice it is quite simple. With the information from the @code{-h} option, we can easily set up our first event counter experiment. We start by using the default set of counters defined for our processor and we use 2 threads: @cartouche @smallexample $ exe=mxv-pthreads.exe $ m=3000 $ n=2000 $ exp=mxv.hwc.def.2.thr.er $ gprofng collect app -O $exp -h auto ./$exe -m $m -n $n -t 2 @end smallexample @end cartouche @IndexSubentry{Options, @code{-h}} @IndexSubentry{Hardware event counters, @code{auto} option} The new option here is @code{-h auto}. The @code{auto} keyword enables hardware event counter profiling and selects the default set of counters defined for this processor. As before, we can display the information, but there is one practical hurdle to take. Unless we like to view all metrics recorded, we would need to know the names of the events that have been enabled. This is tedious and also not portable in case we would like to repeat this experiment on another processor. @IndexSubentry{Hardware event counters, @code{hwc} metric} This is where the special @code{hwc} metric comes very handy. It automatically expands to the active set of events used. With this, it is very easy to display the event counter values. Note that although the regular clock based profiling was enabled, we only want to see the counter values. We also request to see the percentages and limit the output to the first 5 lines: @cartouche @smallexample $ exp=mxv.hwc.def.2.thr.er $ gprofng display text -metrics e.%hwc -limit 5 -functions $exp @end smallexample @end cartouche @smallexample @verbatim Current metrics: e.%cycles:e+%insts:e+%llm:name Current Sort Metric: Exclusive CPU Cycles ( e.%cycles ) Print limit set to 5 Functions sorted by metric: Exclusive CPU Cycles Excl. CPU Excl. Instructions Excl. Last-Level Name Cycles Executed Cache Misses sec. % % % 2.691 100.00 7906475309 100.00 122658983 100.00 2.598 96.54 7432724378 94.01 121745696 99.26 mxv_core 0.035 1.31 188860269 2.39 70084 0.06 erand48_r 0.026 0.95 73623396 0.93 763116 0.62 init_data 0.018 0.66 76824434 0.97 40040 0.03 drand48 @end verbatim @end smallexample As we have seen before, the first few lines echo the settings. This includes a list with the hardware event counters used by default. The table that follows makes it very easy to get an overview where the time is spent and how many of the target events have occurred. As before, we can drill down deeper and see the same metrics at the source line and instruction level. Other than using @code{hwc} in the metrics definitions, nothing has changed compared to the previous examples: @cartouche @smallexample $ exp=mxv.hwc.def.2.thr.er $ gprofng display text -metrics e.hwc -source mxv_core $exp @end smallexample @end cartouche This is the relevant part of the output. Since the lines get very long, we have somewhat modified the lay-out: @smallexample @verbatim Excl. CPU Excl. Excl. Cycles Instructions Last-Level sec. Executed Cache Misses 0. 0 0 32. void __attribute__ ((noinline)) mxv_core(...) 0. 0 0 33. { 0. 0 0 34. for (uint64_t i=...) { 0. 0 0 35. double row_sum = 0.0; ## 1.872 7291879319 88150571 36. for (int64_t j=0; j 0. 0 0 [33] 4021ba: mov 0x8(%rsp),%r10 34. for (uint64_t i=...) { 0. 0 0 [34] 4021bf: cmp %rsi,%rdi 0. 0 0 [34] 4021c2: jbe 0x37 0. 0 0 [34] 4021c4: ret 35. double row_sum = 0.0; 36. for (int64_t j=0; j 2.481 2.481 1233233242 3982327 mxv_core 0.040 0.107 19019012 9003 init_data 0.028 0.052 23023048 15006 erand48_r 0.024 0.024 19019008 9004 __drand48_iterate 0.015 0.067 11011009 2998 drand48 0.008 0.010 0 3002 _int_malloc 0.001 0.001 0 0 brk 0.001 0.002 0 0 sysmalloc 0. 0.001 0 0 __default_morecore @end verbatim @end smallexample @IndexSubentry{Commands, @code{compare ratio}} When using event counters, the values could be very large and it is not easy to compare the numbers. As we will show next, the @code{ratio} feature is very useful when comparing such profiles. To demonstrate this, we have set up another event counter experiment where we would like to compare the number of last level cache miss and the number of branch instructions executed when using a single thread, or two threads. These are the commands used to generate the experiment directories: @cartouche @smallexample $ exe=./mxv-pthreads.exe $ m=3000 $ n=2000 $ exp1=mxv.hwc.comp.1.thr.er $ exp2=mxv.hwc.comp.2.thr.er $ gprofng collect app -O $exp1 -h llm -h br_ins $exe -m $m -n $n -t 1 $ gprofng collect app -O $exp2 -h llm -h br_ins $exe -m $m -n $n -t 2 @end smallexample @end cartouche The following script has been used to get the tables. Due to lay-out restrictions, we have to create two tables, one for each counter. @cartouche @smallexample # Limit the output to 5 lines limit 5 # Define the metrics metrics name:e.llm # Set the comparison to ratio compare ratio functions # Define the metrics metrics name:e.br_ins # Set the comparison to ratio compare ratio functions @end smallexample @end cartouche Note that we print the name of the function first, followed by the counter data. The new element is that we set the comparison mode to @code{ratio}. This divides the data in a column by its counterpart in the reference experiment. This is the command using this script and the two experiment directories as input: @cartouche @smallexample $ gprofng display text -script my-script-comp-counters \ mxv.hwc.comp.1.thr.er \ mxv.hwc.comp.2.thr.er @end smallexample @end cartouche By design, we get two tables, one for each counter: @smallexample @verbatim Functions sorted by metric: Exclusive Last-Level Cache Misses mxv.hwc.comp.1.thr.er mxv.hwc.comp.2.thr.er Name Excl. Last-Level Excl. Last-Level Cache Misses Cache Misses ratio 122709276 x 0.788 mxv_core 121796001 x 0.787 init_data 723064 x 1.055 erand48_r 100111 x 0.500 drand48 60065 x 1.167 Functions sorted by metric: Exclusive Branch Instructions mxv.hwc.comp.1.thr.er mxv.hwc.comp.2.thr.er Name Excl. Branch Excl. Branch Instructions Instructions ratio 1307307316 x 0.997 mxv_core 1235235239 x 0.997 erand48_r 23023033 x 0.957 drand48 20020009 x 0.600 __drand48_iterate 17017028 x 0.882 @end verbatim @end smallexample A ratio less than one in the second column, means that this counter value was smaller than the value from the reference experiment shown in the first column. This kind of presentation of the results makes it much easier to quickly interpret the data. We conclude this section with thread-level event counter overviews, but before we go into this, there is an important metric we need to mention. @IndexSubentry{Hardware event counters, IPC} In case it is known how many instructions and CPU cycles have been executed, the value for the IPC (``Instructions Per Clockycle'') can be computed. @xref{Hardware Event Counters Explained}. This is a derived metric that gives an indication how well the processor is utilized. The inverse of the IPC is called CPI. The @DisplayText{} command automatically computes the IPC and CPI values if an experiment contains the event counter values for the instructions and CPU cycles executed. These are part of the metric list and can be displayed, just like any other metric. @IndexSubentry{Commands, @code{metric_list}} This can be verified through the @code{metric_list} command. If we go back to our earlier experiment with the default event counters, we get the following result. @cartouche @smallexample $ gprofng display text -metric_list mxv.hwc.def.2.thr.er @end smallexample @end cartouche @smallexample @verbatim Current metrics: e.totalcpu:i.totalcpu:e.cycles:e+insts:e+llm:name Current Sort Metric: Exclusive Total CPU Time ( e.totalcpu ) Available metrics: Exclusive Total CPU Time: e.%totalcpu Inclusive Total CPU Time: i.%totalcpu Exclusive CPU Cycles: e.+%cycles Inclusive CPU Cycles: i.+%cycles Exclusive Instructions Executed: e+%insts Inclusive Instructions Executed: i+%insts Exclusive Last-Level Cache Misses: e+%llm Inclusive Last-Level Cache Misses: i+%llm Exclusive Instructions Per Cycle: e+IPC Inclusive Instructions Per Cycle: i+IPC Exclusive Cycles Per Instruction: e+CPI Inclusive Cycles Per Instruction: i+CPI Size: size PC Address: address Name: name @end verbatim @end smallexample Among the other metrics, we see the new metrics for the IPC and CPI listed. In the script below, we use this information and add the IPC and CPI to the metrics to be displayed. We also use a the thread filter to display these values for the individual threads. This is the complete script we have used. Other than a different selection of the metrics, there are no new features. @cartouche @smallexample # Define the metrics metrics e.insts:e.%cycles:e.IPC:e.CPI # Sort with respect to cycles sort e.cycles # Limit the output to 5 lines limit 5 # Get the function overview for all threads functions # Get the function overview for thread 1 thread_select 1 functions # Get the function overview for thread 2 thread_select 2 functions # Get the function overview for thread 3 thread_select 3 functions @end smallexample @end cartouche In the metrics definition on the second line, we explicitly request the counter values for the instructions (@code{e.insts}) and CPU cycles (@code{e.cycles}) executed. These names can be found in output from the @code{metric_list} commad above. In addition to these metrics, we also request the IPC and CPI to be shown. As before, we used the @code{limit} command to control the number of functions displayed. We then request an overview for all the threads, followed by three sets of two commands to select a thread and display the function overview. The script above is used as follows: @cartouche @smallexample $ gprofng display text -script my-script-ipc mxv.hwc.def.2.thr.er @end smallexample @end cartouche This script produces four tables. We list them separately below, and have left out the additional output. The first table shows the accumulated values across the three threads that have been active. @smallexample @verbatim Functions sorted by metric: Exclusive CPU Cycles Excl. Excl. CPU Excl. Excl. Name Instructions Cycles IPC CPI Executed sec. % 7906475309 2.691 100.00 1.473 0.679 7432724378 2.598 96.54 1.434 0.697 mxv_core 188860269 0.035 1.31 2.682 0.373 erand48_r 73623396 0.026 0.95 1.438 0.696 init_data 76824434 0.018 0.66 2.182 0.458 drand48 @end verbatim @end smallexample This shows that IPC of this program is completely dominated by function @code{mxv_core}. It has a fairly low IPC value of 1.43. The next table is for thread 1 and shows the values for the main thread. @smallexample @verbatim Exp Sel Total === === ===== 1 1 3 Functions sorted by metric: Exclusive CPU Cycles Excl. Excl. CPU Excl. Excl. Name Instructions Cycles IPC CPI Executed sec. % 473750931 0.093 100.00 2.552 0.392 188860269 0.035 37.93 2.682 0.373 erand48_r 73623396 0.026 27.59 1.438 0.696 init_data 76824434 0.018 18.97 2.182 0.458 drand48 134442832 0.013 13.79 5.250 0.190 __drand48_iterate @end verbatim @end smallexample Although this thread hardly uses any CPU cycles, the overall IPC of 2.55 is not all that bad. Last, we show the tables for threads 2 and 3: @smallexample @verbatim Exp Sel Total === === ===== 1 2 3 Functions sorted by metric: Exclusive CPU Cycles Excl. Excl. CPU Excl. Excl. Name Instructions Cycles IPC CPI Executed sec. % 3716362189 1.298 100.00 1.435 0.697 3716362189 1.298 100.00 1.435 0.697 mxv_core 0 0. 0. 0. 0. collector_root 0 0. 0. 0. 0. driver_mxv Exp Sel Total === === ===== 1 3 3 Functions sorted by metric: Exclusive CPU Cycles Excl. Excl. CPU Excl. Excl. Name Instructions Cycles IPC CPI Executed sec. % 3716362189 1.300 100.00 1.433 0.698 3716362189 1.300 100.00 1.433 0.698 mxv_core 0 0. 0. 0. 0. collector_root 0 0. 0. 0. 0. driver_mxv @end verbatim @end smallexample It is seen that both execute the same number of instructions and take about the same number of CPU cycles. As a result, the IPC is the same for both threads. @c -- A new node -------------------------------------------------------------- @c TBD @node Additional Features @c TBD @section Additional Features @c ---------------------------------------------------------------------------- @c -- A new node -------------------------------------------------------------- @c TBD @node More Filtering Capabilities @c TBD @subsection More Filtering Capabilities @c ---------------------------------------------------------------------------- @c TBD Cover @code{samples} and @code{seconds} @c -- A new node -------------------------------------------------------------- @node Java Profiling @section Java Profiling @c ---------------------------------------------------------------------------- @IndexSubentry{Java profiling, @code{-j on/off}} The @CollectApp{} command supports Java profiling. The @code{-j on} option can be used for this, but since this feature is enabled by default, there is no need to set this explicitly. Java profiling may be disabled through the @code{-j off} option. The program is compiled as usual and the experiment directory is created similar to what we have seen before. The only difference with a C/C++ application is that the program has to be explicitly executed by java. For example, this is how to generate the experiment data for a Java program that has the source code stored in file @code{Pi.java}: @cartouche @smallexample $ javac Pi.java $ gprofng collect app -j on -O pi.demo.er java Pi < pi.in @end smallexample @end cartouche Regarding which java is selected to generate the data, @ToolName{} first looks for the JDK in the path set in either the @IndexSubentry{Java profiling, @code{JDK_HOME}} @code{JDK_HOME} environment variable, or in the @IndexSubentry{Java profiling, @code{JAVA_PATH}} @code{JAVA_PATH} environment variable. If neither of these variables is set, it checks for a JDK in the search path (set in the PATH environment variable). If there is no JDK in this path, it checks for the java executable in @code{/usr/java/bin/java}. In case additional options need to be passed on to the JVM, the @IndexSubentry{Java profiling, @code{-J }} @code{-J } option can be used. The string with the option(s) has to be delimited by quotation marks in case there is more than one argument. The @DisplayText{} command may be used to view the performance data. There is no need for any special options and the same commands as previously discussed are supported. @IndexSubentry{Commands, @code{viewmode}} @IndexSubentry{Java profiling, different view modes} The @code{viewmode} command @xref{The Viewmode} is very useful to examine the call stacks. For example, this is how one can see the native call stacks. For lay-out purposes we have restricted the list to the first five entries: @cartouche @smallexample $ gprofng display text -limit 5 -viewmode machine -calltree pi.demo.er @end smallexample @end cartouche @smallexample @verbatim Print limit set to 5 Viewmode set to machine Functions Call Tree. Metric: Attributed Total CPU Time Attr. Name Total CPU sec. 1.381 +- 1.171 +-Pi.calculatePi(double) 0.110 +-collector_root 0.110 | +-JavaMain 0.070 | +-jni_CallStaticVoidMethod @end verbatim @end smallexample @noindent Note that the selection of the viewmode is echoed in the output. @c -- A new node -------------------------------------------------------------- @c TBD @node Summary of Options and Commands @c TBD @chapter Summary of Options and Commands @c ---------------------------------------------------------------------------- @c -- A new node -------------------------------------------------------------- @node Terminology @chapter Terminology Throughout this manual, certain terminology specific to profiling tools, or @ToolName{}, or even to this document only, is used. In this chapter we explain this terminology in detail. @menu * The Program Counter:: What is a Program Counter? * Inclusive and Exclusive Metrics:: An explanation of inclusive and exclusive metrics. * Metric Definitions:: Definitions associated with metrics. * The Viewmode:: Select the way call stacks are presented. * The Selection List:: How to define a selection. * Load Objects and Functions:: The components in an application. * The Concept of a CPU in @ProductName{}:: The definition of a CPU. * Hardware Event Counters Explained:: What are event counters? * apath:: Our generic definition of a path. @end menu @c ---------------------------------------------------------------------------- @node The Program Counter @section The Program Counter @c ---------------------------------------------------------------------------- @cindex PC @cindex Program Counter The @emph{Program Counter}, or PC for short, keeps track where program execution is. The address of the next instruction to be executed is stored in a special purpose register in the processor, or core. @cindex Instruction pointer The PC is sometimes also referred to as the @emph{instruction pointer}, but we will use Program Counter or PC throughout this document. @c ---------------------------------------------------------------------------- @node Inclusive and Exclusive Metrics @section Inclusive and Exclusive Metrics @c ---------------------------------------------------------------------------- In the remainder, these two concepts occur quite often and for lack of a better place, they are explained here. @cindex Inclusive metric The @emph{inclusive} value for a metric includes all values that are part of the dynamic extent of the target function. For example if function @code{A} calls functions @code{B} and @code{C}, the inclusive CPU time for @code{A} includes the CPU time spent in @code{B} and @code{C}. @cindex Exclusive metric In contrast with this, the @emph{exclusive} value for a metric is computed by excluding the metric values used by other functions called. In our imaginary example, the exclusive CPU time for function @code{A} is the time spent outside calling functions @code{B} and @code{C}. @cindex Leaf function In case of a @emph{leaf function}, the inclusive and exclusive values for the metric are the same since by definition, it is not calling any other function(s). Why do we use these two different values? The inclusive metric shows the most expensive path, in terms of this metric, in the application. For example, if the metric is cache misses, the function with the highest inclusive metric tells you where most of the cache misses come from. Within this branch of the application, the exclusive metric points to the functions that contribute and help to identify which part(s) to consider for further analysis. @c ---------------------------------------------------------------------------- @node Metric Definitions @section Metric Definitions @c ---------------------------------------------------------------------------- The metrics to be shown are highly customizable. In this section we explain the definitions associated with metrics. @IndexSubentry{Commands, @code{metrics}} The @code{metrics} command takes a colon (:) separated list with special keywords. This keyword consists of the following three fields: @code{}@code{}@code{}. @cindex Flavor field @cindex Visibility field @cindex Metric name field The @emph{} field is either an @code{e} for ``exclusive'', or @code{i} for ``inclusive''. The @code{} field is the name of the metric request. The @emph{} field consists of one ore more characters from the following table: @table @code @item . Show the metric as time. This applies to timing metrics and hardware event counters that measure cycles. Interpret as @code{+} for other metrics. @item % Show the metric as a percentage of the total value for this metric. @item + Show the metric as an absolute value. For hardware event counters this is the event count. Interpret as @code{.} for timing metrics. @item | Do not show any metric value. Cannot be used with other visibility characters. @end table @c ---------------------------------------------------------------------------- @node The Viewmode @section The Viewmode @cindex Viewmode @IndexSubentry{Commands, @code{viewmode}} There are different ways to view a call stack in Java. In @ToolName{}, this is called the @emph{viewmode} and the setting is controlled through a command with the same name. The @code{viewmode} command takes one of the following keywords: @table @code @item user This is the default and shows the Java call stacks for Java threads. No call stacks for any housekeeping threads are shown. The function list contains a function @IndexSubentry{Java profiling, @code{}} @code{} that represents the aggregated time from non-Java threads. When the JVM software does not report a Java call stack, time is reported against the function @IndexSubentry{Java profiling, @code{}} @code{}. @item expert Show the Java call stacks for Java threads when the Java code from the user is executed and machine call stacks when JVM code is executed, or when the JVM software does not report a Java call stack. Show the machine call stacks for housekeeping threads. @item machine Show the actual native call stacks for all threads. @end table @c ---------------------------------------------------------------------------- @c ---------------------------------------------------------------------------- @node The Selection List @section The Selection List @c ---------------------------------------------------------------------------- @cindex Selection list @cindex List specification Several commands allow the user to specify a subset of a list. For example, to select specific threads from all the threads that have been used when conducting the experiment(s). Such a selection list (or ``list'' in the remainder of this section) can be a single number, a contiguous range of numbers with the start and end numbers separated by a hyphen (@code{-}), a comma-separated list of numbers and ranges, or the @code{all} keyword. Lists must not contain spaces. Each list can optionally be preceded by an experiment list with a similar format, separated from the list by a colon (:). If no experiment list is included, the list applies to all experiments. Multiple lists can be concatenated by separating the individual lists by a plus sign. These are some examples of various filters using a list: @table @code @item thread_select 1 Select thread 1 from all experiments. @item thread_select all:1 Select thread 1 from all experiments. @item thread_select 1:1+2:2 Select thread 1 from experiment 1 and thread 2 from experiment 2. @item cpu_select all:1,3,5 Selects cores 1, 3, and 5 from all experiments. @item cpu_select 1,2:all Select all cores from experiments 1 and 2, as listed by the @code{by exp_list} command. @end table @c ---------------------------------------------------------------------------- @node Load Objects and Functions @section Load Objects and Functions @c ---------------------------------------------------------------------------- An application consists of various components. The source code files are compiled into object files. These are then glued together at link time to form the executable. During execution, the program may also dynamically load objects. @cindex Load object A @emph{load object} is defined to be an executable, or shared object. A shared library is an example of a load object in @ToolName{}. Each load object, contains a text section with the instructions generated by the compiler, a data section for data, and various symbol tables. All load objects must contain an @cindex ELF ELF symbol table, which gives the names and addresses of all the globally known functions in that object. Load objects compiled with the -g option contain additional symbolic information that can augment the ELF symbol table and provide information about functions that are not global, additional information about object modules from which the functions came, and line number information relating addresses to source lines. The term @cindex Function @emph{function} is used to describe a set of instructions that represent a high-level operation described in the source code. The term also covers methods as used in C++ and in the Java programming language. In the @ToolName{} context, functions are provided in source code format. Normally their names appear in the symbol table representing a set of addresses. @cindex Program Counter @cindex PC If the Program Counter (PC) is within that set, the program is executing within that function. In principle, any address within the text segment of a load object can be mapped to a function. Exactly the same mapping is used for the leaf PC and all the other PCs on the call stack. Most of the functions correspond directly to the source model of the program, but there are exceptions. This topic is however outside of the scope of this guide. @c ---------------------------------------------------------------------------- @node The Concept of a CPU in @ProductName{} @section The Concept of a CPU in @ProductName{} @c ---------------------------------------------------------------------------- @cindex CPU In @ProductName{}, there is the concept of a CPU. Admittedly, this is not the best word to describe what is meant here and may be replaced in the future. The word CPU is used in many of the displays. In the context of @ProductName{}, it is meant to denote a part of the processor that is capable of executing instructions and with its own state, like the program counter. For example, on a contemporary processor, a CPU could be a core. In case hardware threads are supported within a core, it could be one of those hardware threads. @c ---------------------------------------------------------------------------- @node Hardware Event Counters Explained @section Hardware Event Counters Explained @c ---------------------------------------------------------------------------- @IndexSubentry{Hardware event counters, description} For quite a number of years now, many microprocessors have supported hardware event counters. On the hardware side, this means that in the processor there are one or more registers dedicated to count certain activities, or ``events''. Examples of such events are the number of instructions executed, or the number of cache misses at level 2 in the memory hierarchy. While there is a limited set of such registers, the user can map events onto them. In case more than one register is available, this allows for the simultaenous measurement of various events. A simple, yet powerful, example is to simultaneously count the number of CPU cycles and the number of instructions excuted. These two numbers can then be used to compute the @cindex IPC @emph{IPC} value. IPC stands for ``Instructions Per Clockcycle'' and each processor has a maximum. For example, if this maximum number is 2, it means the processor is capable of executing two instructions every clock cycle. Whether this is actually achieved, depends on several factors, including the instruction characteristics. However, in case the IPC value is well below this maximum in a time critical part of the application and this cannot be easily explained, further investigation is probably warranted. @cindex CPI A related metric is called @emph{CPI}, or ``Clockcycles Per Instruction''. It is the inverse of the CPI and can be compared against the theoretical value(s) of the target instruction(s). A significant difference may point at a bottleneck. One thing to keep in mind is that the value returned by a counter can either be the number of times the event occured, or a CPU cycle count. In case of the latter it is possible to convert this number to time. @IndexSubentry{Hardware event counters, variable CPU frequency} This is often easier to interpret than a simple count, but there is one caveat to keep in mind. The CPU frequency may not have been constant while the experimen was recorded and this impacts the time reported. These event counters, or ``counters'' for short, provide great insight into what happens deep inside the processor. In case higher level information does not provide the insight needed, the counters provide the information to get to the bottom of a performance problem. There are some things to consider though. @itemize @bullet @item The event definitions and names vary across processors and it may even happen that some events change with an update. Unfortunately and this is luckily rare, there are sometimes bugs causing the wrong count to be returned. @IndexSubentry{Hardware event counters, alias name} In @ToolName{}, some of the processor specific event names have an alias name. For example @code{insts} measures the instructions executed. These aliases not only makes it easier to identify the functionality, but also provide portability of certain events across processors. @item Another complexity is that there are typically many events one can monitor. There may up to hundreds of events available and it could require several experiments to zoom in on the root cause of a performance problem. @item There may be restrictions regarding the mapping of event(s) onto the counters. For example, certain events may be restricted to specific counters only. As a result, one may have to conduct additional experiments to cover all the events of interest. @item The names of the events may also not be easy to interpret. In such cases, the description can be found in the architecture manual for the processor. @end itemize Despite these drawbacks, hardware event counters are extremely useful and may even turn out to be indispensable. @c ---------------------------------------------------------------------------- @node apath @section What is ? @c ---------------------------------------------------------------------------- In most cases, @ToolName{} shows the absolute pathnames of directories. These tend to be rather long, causing display issues in this document. Instead of wrapping these long pathnames over multiple lines, we decided to represent them by the @code{} symbol, which stands for ``an absolute pathname''. Note that different occurrences of @code{} may represent different absolute pathnames. @c -- A new node -------------------------------------------------------------- @node Other Document Formats @chapter Other Document Formats @c ---------------------------------------------------------------------------- This document is written in Texinfo and the source text is made available as part of the binutils distribution. The file name is @code{gprofng.texi} and can be found in subdirectory @code{doc} under directory @code{gprofng} in the top level directory. This file can be used to generate the document in the @code{info}, @code{html}, and @code{pdf} formats. The default installation procedure creates a file in the @code{info} format and stores it in the documentation section of binutils. The probably easiest way to generate a different format from this Texinfo document is to go to the distribution directory that was created when the tools were built. This is either the default distribution directory, or the one that has been set with the @code{--prefix} option as part of the @code{configure} command. In this example we symbolize this location with @code{}. The make file called @code{Makefile} in directory @code{/gprofng/doc} supports several commands to generate this document in different formats. We recommend to use these commands. They create the file(s) and install it in the documentation directory of binutils, which is @code{/share/doc} in case @code{html} or @code{pdf} is selected and @code{/share/info} for the file in the @code{info} format. To generate this document in the requested format and install it in the documentation directory, the commands below should be executed. In this notation, @code{} is one of @code{info}, @code{html}, or @code{pdf}: @smallexample @verbatim $ cd /gprofng/doc $ make install- @end verbatim @end smallexample @noindent Some things to note: @itemize @item For the @code{pdf} file to be generated, the @cindex TeX TeX document formatting software is required and the relevant commmands need to be included in the search path. An example of a popular TeX implementation is @emph{TexLive}. It is beyond the scope of this document to go into the details of installing and using TeX, but it is well documented elsewhere. @item Instead of generating a single file in the @code{html} format, it is also possible to create a directory with individual files for the various chapters. To do so, remove the use of @code{--no-split} in variable @code{MAKEINFOHTML} in the make file in the @code{doc} directory. @item The make file also supports commands to only generate the file in the desired format and not move them to the documentation directory. This is accomplished through the @code{make } command. @end itemize @ifnothtml @node Index @unnumbered Index @printindex cp @end ifnothtml @bye