<p>If your CPU is really old, it may not have SSE2. In this case, you need
to <aclass="reference internal"href="#translate">translate</a> yourself with the option <ttclass="docutils literal"><spanclass="pre">--jit-backend=x86-without-sse2</span></tt>.</p>
<p>A special safe version. Read the docs about <aclass="reference external"href="features.html#sandboxing">sandboxing</a>. These
binaries work on 32-bit <aclass="reference external"href="http://en.wikipedia.org/wiki/IA-32">x86 (IA-32)</a> CPUs as well as <aclass="reference external"href="http://en.wikipedia.org/wiki/X86-64">x86-64</a> CPUs
<p>It is also possible to <aclass="reference internal"href="#translate">translate</a> a version that includes both
<li>No JIT: A version without the JIT. Consumes a bit less memory
and may be faster on short-running scripts.</li>
<li>Stackless: Provides <aclass="reference external"href="http://www.stackless.com/">Stackless</a> extensions, as well as <aclass="reference external"href="http://codespeak.net/svn/greenlet/trunk/doc/greenlet.txt">greenlets</a>.
It is not possible right now to combine Stackless features with the JIT.</li>
<li>Sandboxing: A special safe version. Read the docs about <aclass="reference external"href="features.html#sandboxing">sandboxing</a>.
(It is also possible to <aclass="reference internal"href="#translate">translate</a> a version that includes both
sandboxing and the JIT compiler, although as the JIT is relatively
...
...
@@ -88,14 +87,4 @@
sandboxing and the JIT compiler, although as the JIT is relatively
complicated, this reduce a bit the level of confidence we can put in
the result.</p>
<p>The Windows need testing and careful reviewing; <aclass="reference external"href="contact.html">contact us</a>!</p>
<p>Provides <aclass="reference external"href="http://www.stackless.com/">Stackless</a> extensions, as well as <aclass="reference external"href="http://codespeak.net/svn/greenlet/trunk/doc/greenlet.txt">greenlets</a>. These
binaries work on 32-bit <aclass="reference external"href="http://en.wikipedia.org/wiki/IA-32">x86 (IA-32)</a> CPUs as well as <aclass="reference external"href="http://en.wikipedia.org/wiki/X86-64">x86-64</a> CPUs
complicated, this reduces a bit the level of confidence we can put in
the result.)</li>
</ul>
...
...
@@ -101,6 +90,8 @@
</ul>
<p>It is not possible right now to combine Stackless features with the JIT.</p>
<p>These versions are not officially part of the release 1.4, which focuses
on the JIT. You can find prebuilt binaries for them on our
<aclass="reference external"href="http://buildbot.pypy.org/nightly/trunk/">nightly build</a> or with the <aclass="reference external"href="http://pypy.org/download/">older releases</a>, or <aclass="reference internal"href="#translate">translate</a> them yourself.</p>
</div>
<divclass="section"id="installing">
<h1>Installing</h1>
<p>All versions are packaged in a <ttclass="docutils literal">tar.bz2</tt> or <ttclass="docutils literal">zip</tt> file. When
...
...
@@ -103,15 +94,13 @@
</div>
<divclass="section"id="installing">
<h1>Installing</h1>
<p>All versions are packaged in a <ttclass="docutils literal">tar.bz2</tt> or <ttclass="docutils literal">zip</tt> file. When
uncompressed, they run in-place. On Linux or Mac OS/X, they can also be
installed by manually moving the files:</p>
<preclass="literal-block">
/usr/bin/pypy # or pypy-nojit etc.
/usr/share/pypy-1.4/lib-python/*
/usr/share/pypy-1.4/pypy/*
</pre>
<p>You can also install it to <ttclass="docutils literal">/usr/local/bin</tt> and <ttclass="docutils literal">/usr/local/share</tt>.</p>
uncompressed, they run in-place. For now you can uncompress them
either somewhere in your home directory or, say, in <ttclass="docutils literal">/opt</tt>, and
if you want, put a symlink from somewhere like
<ttclass="docutils literal">/usr/local/bin/pypy</tt> to <ttclass="docutils literal"><spanclass="pre">/path/to/pypy-1.4/bin/pypy</span></tt>. Do
not move or copy the executable <ttclass="docutils literal">pypy</tt> outside the tree – put
a symlink to it, otherwise it will not find its libraries.</p>
</div>
<divclass="section"id="building-from-source">
<spanid="translate"></span><h1>Building from source</h1>
...
...
@@ -127,9 +116,6 @@
<preclass="literal-block">
svn co http://codespeak.net/svn/pypy/trunk pypy-trunk
</pre>
<p>Windows users trying the zip file are expected to use MSVC.
The trunk version contains fixes to support MinGW32. More
information on <aclass="reference external"href="http://codespeak.net/pypy/trunk/pypy/doc/windows.html">our dev site</a>.</p>
</li>
<li><pclass="first">Make sure you installed the dependencies. See the list <aclass="reference external"href="http://codespeak.net/pypy/dist/pypy/doc/getting-started-python.html#translating-the-pypy-python-interpreter">here</a>.</p>
</li>
...
...
@@ -149,8 +135,13 @@
</pre>
</li>
<li><pclass="first">Enjoy Mandelbrot <ttclass="docutils literal"><spanclass="pre">:-)</span></tt> It takes on the order of one hour to
finish the translation, and 1.3 GB of RAM on a 32-bit system.
(Do not start a translation on a machine with 1GB or less!)</p>
finish the translation, and 1.2 GB of RAM on a 32-bit system
and 2.4 GB on 64-bit systems. (Do not start a translation on a
machine with insufficient RAM! It will just swap forever.)</p>
</li>
<li><pclass="first">Once you have a PyPy, it is recommended to use it to do further
translations, instead of using CPython. It is twice as fast,
but uses a bit more RAM (1.7-2.0 GB on 32-bit, 3.0 GB on 64-bit).</p>