Profiling PHP Scripts

Xdebug's built-in profiler allows you to find bottlenecks in your script and visualize those with an external tool such as KCacheGrind or WinCacheGrind.

Introduction

Xdebug's Profiler is a powerful tool that gives you the ability to analyze your PHP code and determine bottlenecks or generally see which parts of your code are slow and could use a speed boost. Since Xdebug 2.6, the profiler also collects information about how much memory is being used, and which functions aGnd methods increased memory usage.

The profiler in Xdebug outputs profiling information in the form of a Cachegrind compatible file. This allows you to use the excellent KCacheGrind tool (Linux, KDE) to analyse your profiling data. If you are on Linux you can install KCacheGrind with your favourite package manager.

If you are on Windows, there are precompiled QCacheGrind binaries available. (QCacheGrind is KCacheGrind without KDE bindings).

If you are on Mac OSX, there are instructions on how to build QCacheGrind too.

Users of Windows can alternatively use WinCacheGrind. The functionality is different from KCacheGrind so the section that documents the use of KCacheGrind on this page doesn't apply to this program. WinCacheGrind currently does not support the file and function compression for cachegrind files that Xdebug 2.3 introduces yet.

There is also an alternative profile information presentation tool called xdebugtoolkit, a web based front-end called Webgrind, and a Java based tool called XCallGraph.

In case you can not use KDE (or do not want to use KDE) the kcachegrind package also comes with a perl script "ct_annotate" which produces ASCII output from the profiler trace files.

Starting The Profiler

Profiling is enabled by setting the xdebug.profiler_enable setting to 1 in php.ini. This instructs Xdebug to start writing profiling information into the dump directory configured with the xdebug.profiler_output_dir directive. The name of the generated file always starts with "cachegrind.out." and ends with either the PID (process ID) of the PHP (or Apache) process or the crc32 hash of the directory containing the initially debugged script. Make sure you have enough space in your xdebug.profiler_output_dir as the amount of information generated by the profiler can be enormous for complex scripts, for example up to 500MB for a complex application like eZ Publish.

You can also selectively enable the profiler with the xdebug.profiler_enable_trigger setting set to 1. If it is set to 1, then you can enable the profiler by using a GET/POST or COOKIE variable of the name XDEBUG_PROFILE. The FireFox 2 extension that can be used to enable the debugger (see HTTP Debug Sessions) can also be used with this setting. In order for the trigger to work properly, xdebug.profiler_enable needs to be set to 0.

From Xdebug 2.6 onwards, Xdebug adds the HTTP header X-Xdebug-Profile-Filename to a request that is being profiled. This header contains the name of the file that holds the profiling information for that request.

Analysing Profiles

After a profile information file has been generated you can open it with KCacheGrind:

Once the file is opened you have plenty of information available in the different panes of KCacheGrind. On the left side you find the "Flat Profile" pane showing all functions in your script sorted by time spend in this function, and all its children. The second column "Self" shows the time spend in this function (without its children), the third column "Called" shows how often a specific function was called and the last column "Function" shows the name of the function. Xdebug changes internal PHP function names by prefixing the function name with "php::" and include files are handled in a special way too. Calls to include (and include_one, require and require_once) are followed by "::" and the filename of the included file. In the screenshot on the left you can see this for "include::/home/httpd/ez_34/v..." and an example of an internal PHP function is "php::mysql_query". The numbers in the first two columns can be either percentages of the full running time of the script (like in the example) or absolute time (1 unit is 1/1.000.000th of a second). You can switch between the two modes with the button you see on the right.

The pane on the right contains an upper and lower pane. The upper one shows information about which functions called the current selected function ("eztemplatedesignresource->executecompiledtemplate in the screenshot). The lower pane shows information about the functions that the current selected function called.

The "Cost" column in the upper pane shows the time spend in the current selected function while being called from the function in the list. The numbers in the Cost column added up will always be 100%. The "Cost" column in the lower pane shows the time spend while calling the function from the list. While adding the numbers in this list up, you will most likely never reach 100% as the selected function itself will also takes time to execute.

The "All Callers" and "All Calls" tabs show not only the direct call from which the function was called respectively all directly made function calls but also function calls made more levels up and down. The upper pane in the screenshot on the left shows all functions calling the current selected one, both directly and indirectly with other functions inbetween them on the stack. The "Distance" column shows how many function calls are between the listed and the current selected one (-1). If there are different distances between two functions, it is shown as a range (for example "5-24"). The number in parentheses is the median distance. The lower pane is similar except that it shows information on functions called from the current selected one, again either direct or indirect.