forked from OSchip/llvm-project
233 lines
12 KiB
HTML
233 lines
12 KiB
HTML
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Transitional//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-transitional.dtd">
|
|
<html xmlns="http://www.w3.org/1999/xhtml">
|
|
<head>
|
|
<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1" />
|
|
<link href="style.css" rel="stylesheet" type="text/css" />
|
|
<title>Testing LLDB</title>
|
|
</head>
|
|
<body>
|
|
<div class="www_title">
|
|
The <strong>LLDB</strong> Debugger
|
|
</div>
|
|
|
|
<div id="container">
|
|
<div id="content">
|
|
|
|
<!--#include virtual="sidebar.incl"-->
|
|
|
|
<div id="middle">
|
|
<div class="post">
|
|
<h1 class="postheader">Testing LLDB</h1>
|
|
<div class="postcontent">
|
|
<p>
|
|
The LLDB test suite consists of Python scripts located under the
|
|
<tt>test</tt> directory. Each script contains a number of test cases and is usually
|
|
accompanied by a C (C++, ObjC, etc.) source file. Each test first compiles the
|
|
source file and then uses LLDB to debug the resulting executable. The tests verify
|
|
both the LLDB command line interface and the scripting API.
|
|
</p>
|
|
</div>
|
|
<h1 class="postheader">Running tests</h1>
|
|
<div class="postcontent">
|
|
<h2>Running the full test suite</h2>
|
|
<p>
|
|
<strong>Windows Note</strong>: In the examples that follow, any invocations of <code>python</code>
|
|
should be replaced with <code>python_d</code>, the debug interpreter, when running the test
|
|
suite against a debug version of LLDB.
|
|
</p>
|
|
<p>
|
|
The easiest way to run the LLDB test suite is to use the <tt>check-lldb</tt> build
|
|
target. By default, the <tt>check-lldb</tt> target builds the test programs with
|
|
the same compiler that was used to build LLDB. To build the tests with a different
|
|
compiler, you can set the <strong>LLDB_TEST_C_COMPILER</strong> or the <strong>LLDB_TEST_CXX_COMPILER</strong> CMake variables.
|
|
It is possible to customize the architecture of the test binaries and compiler used by appending -A
|
|
and -C options respectively to the CMake variable <strong>LLDB_TEST_USER_ARGS</strong>. For
|
|
example, to test LLDB against 32-bit binaries
|
|
built with a custom version of clang, do:
|
|
</p>
|
|
<code>
|
|
<br />> cmake -DLLDB_TEST_USER_ARGS="-A i386 -C /path/to/custom/clang" -G Ninja
|
|
<br />> ninja check-lldb
|
|
</code>
|
|
<p>Note that multiple -A and -C flags can be specified to <tt>LLDB_TEST_USER_ARGS</tt>.</p>
|
|
<p>Note that on NetBSD you must export <tt>LD_LIBRARY_PATH=$PWD/lib</tt> in your environment. This is due to lack of
|
|
the <tt>$ORIGIN</tt> linker feature.</p>
|
|
<h2>Running a specific test or set of tests</h2>
|
|
<p>
|
|
In addition to running all the LLDB test suites with the "check-lldb" CMake target above, it is possible to
|
|
run individual LLDB tests. For example, to run the test cases defined in TestInferiorCrashing.py, run:
|
|
</p>
|
|
<code>
|
|
<br />> cd $lldb/test
|
|
<br />> python dotest.py --executable <path-to-lldb> -p TestInferiorCrashing.py ../packages/Python/lldbsuite/test
|
|
</code>
|
|
<p>
|
|
If the test is not specified by name (e.g. if you leave the <code>-p</code> argument off), LLDB will run all tests in
|
|
that directory:
|
|
</p>
|
|
<code>
|
|
<br />> python dotest.py --executable <path-to-lldb> functionalities/data-formatter
|
|
</code>
|
|
<p>
|
|
Many more options that are available. To see a list of all of them, run:
|
|
</p>
|
|
<code>
|
|
> python dotest.py -h
|
|
</code>
|
|
|
|
<p>
|
|
The dotest.py script runs tests in parallel by default.
|
|
To disable the parallel test running feature, use the
|
|
<code>--no-multiprocess</code> flag. The number of
|
|
concurrent tests is controlled by
|
|
the <code>LLDB_TEST_THREADS</code> environment variable
|
|
or the <code>--threads</code> command line parameter.
|
|
The default value is the number of CPU cores on your
|
|
system.
|
|
</p>
|
|
<p>
|
|
The parallel test running feature will handle an
|
|
additional <code>--test-subdir SUBDIR</code> arg. When
|
|
specified, SUBDIR is relative to the root test directory
|
|
and will limit all parallel test running to that
|
|
subdirectory's tree of tests.
|
|
</p>
|
|
<p>
|
|
The parallel test runner will run all tests within a
|
|
given directory serially, but will run multiple
|
|
directories concurrently. Thus, as a test writer, we
|
|
provide serialized test run semantics within a
|
|
directory. Note child directories are considered
|
|
entirely separate, so two child directories could be
|
|
running in parallel with a parent directory.
|
|
</p>
|
|
|
|
<h2>Running the test-suite remotely</h2>
|
|
|
|
<p>
|
|
Running the test-suite remotely is similar to the process of running a local test
|
|
suite, but there are two things to have in mind:
|
|
</p>
|
|
<ul>
|
|
<li>
|
|
You must have the <code>lldb-server</code> running on the remote system, ready to
|
|
accept multiple connections. For more information on how to setup remote
|
|
debugging see the <a href="remote.html">Remote debugging</a> page.
|
|
</li>
|
|
<li>
|
|
You must tell the test-suite how to connect to the remote system. This is
|
|
achieved using the <code>--platform-name</code>, <code>--platform-url</code> and
|
|
<code>--platform-working-dir</code> parameters to <code>dotest.py</code>. These
|
|
parameters correspond to the <code>platform select</code> and <code>platform
|
|
connect</code> LLDB commands. You will usually also need to specify the compiler and
|
|
architecture for the remote system.
|
|
</li>
|
|
</ul>
|
|
<p>
|
|
Currently, running the remote test suite is supported only with
|
|
<code>dotest.py</code> (or <code>dosep.py</code> with a single thread), but we
|
|
expect this issue to be addressed in the near future.
|
|
</p>
|
|
|
|
</div>
|
|
<div class="postfooter"></div>
|
|
<h1 class="postheader">Debugging test failures</h1>
|
|
<div class="postcontent">
|
|
<h2>Non-Windows platforms</h2>
|
|
<p>
|
|
On non-Windows platforms, you can use the <code>-d</code> option to <code>dotest.py</code> which will cause the script to wait
|
|
for a while until a debugger is attached.
|
|
</p>
|
|
<h2>Windows</h2>
|
|
<p>
|
|
On Windows, it is strongly recommended to use <a href="https://github.com/Microsoft/PTVS/releases">Python Tools for Visual Studio</a>
|
|
for debugging test failures. It can seamlessly step between native and managed code, which is very helpful when you need to step
|
|
through the test itself, and then into the LLDB code that backs the operations the test is performing. A quick guide to getting
|
|
started with PTVS is as follows:
|
|
<ul>
|
|
<li>Install PTVS</li>
|
|
<li>
|
|
Create a Visual Studio Project for the Python code.
|
|
<ul>
|
|
<li>Go to File -> New -> Project -> Python -> From Existing Python Code.</li>
|
|
<li>Choose <code>llvm/tools/lldb</code> as the directory containing the Python code.</li>
|
|
<li>
|
|
When asked where to save the <code>.pyproj</code> file, choose the folder <code>llvm/tools/lldb/pyproj</code>.
|
|
This is a special folder that is ignored by the <code>.gitignore</code> file, since it is not checked in.
|
|
</li>
|
|
</ul>
|
|
</li>
|
|
<li>Set <code>test/dotest.py</code> as the startup file</li>
|
|
<li>
|
|
Make sure there is a Python Environment installed for your distribution. For example, if you installed Python to
|
|
<code>C:\Python35</code>, PTVS needs to know that this is the interpreter you want to use for running the test suite.
|
|
<ul>
|
|
<li>Go to Tools -> Options -> Python Tools -> Environment Options</li>
|
|
<li>Click Add Environment, and enter <code>Python 3.5 Debug</code> for the name. Fill out the values correctly.</li>
|
|
</ul>
|
|
</li>
|
|
<li>
|
|
Configure the project to use this debug interpreter.
|
|
<ul>
|
|
<li>Right click the Project node in Solution Explorer</li>
|
|
<li>In the <code>General</code> tab, Make sure <code>Python 3.5 Debug</code> is the selected Interpreter.</li>
|
|
<li>In <code>Debug/Search Paths</code>, enter the path to your <code>ninja/lib/site-packages</code> directory.</li>
|
|
<li>
|
|
In <code>Debug/Environment Variables</code>, enter<br/>
|
|
<code>VCINSTALLDIR=C:\Program Files (x86)\Microsoft Visual Studio 14.0\VC\</code>
|
|
</li>
|
|
<li>
|
|
If you want to enabled mixed mode debugging, check <code>Enable native code debugging</code> (this slows down debugging,
|
|
so enable it only on an as-needed basis.)
|
|
</li>
|
|
</ul>
|
|
</li>
|
|
<li>
|
|
Set the command line for the test suite to run.
|
|
<ul>
|
|
<li>Right click the project in solution explorer and choose the <code>Debug</code> tab.</li>
|
|
<li>Enter the arguments to <code>dotest.py</code>. Note you must add <code>--no-multiprocess</code></li>
|
|
<li>
|
|
Example command options:
|
|
<code>
|
|
<br/># quiet mode
|
|
<br/>-q
|
|
<br />--arch=i686
|
|
<br /># Path to debug lldb.exe
|
|
<br />--executable D:/src/llvmbuild/ninja/bin/lldb.exe
|
|
<br /># Directory to store log files
|
|
<br />-s D:/src/llvmbuild/ninja/lldb-test-traces
|
|
<br />-u CXXFLAGS -u CFLAGS
|
|
<br /># If a test crashes, show JIT debugging dialog.
|
|
<br />--enable-crash-dialog
|
|
<br /># Path to release clang.exe
|
|
<br />-C d:\src\llvmbuild\ninja_release\bin\clang.exe
|
|
<br /># Path to the particular test you want to debug.
|
|
<br />-p TestPaths.py
|
|
<br /># Root of test tree
|
|
<br />D:\src\llvm\tools\lldb\packages\Python\lldbsuite\test
|
|
<br /># Required in order to be able to debug the test.
|
|
<br />--no-multiprocess
|
|
</code>
|
|
</li>
|
|
<li>
|
|
As copy-pastable command line:<br/>
|
|
<code>
|
|
-q --arch=i686 --executable D:/src/llvmbuild/ninja/bin/lldb.exe -s D:/src/llvmbuild/ninja/lldb-test-traces
|
|
-u CXXFLAGS -u CFLAGS --enable-crash-dialog -C d:\src\llvmbuild\ninja_release\bin\clang.exe
|
|
-p TestPaths.py D:\src\llvm\tools\lldb\packages\Python\lldbsuite\test --no-multiprocess
|
|
</code>
|
|
</li>
|
|
</ul>
|
|
</li>
|
|
</ul>
|
|
</p>
|
|
</div>
|
|
<div class="postfooter"></div>
|
|
</div>
|
|
</div>
|
|
</div>
|
|
</div>
|
|
</body>
|
|
</html>
|