ref/distrib/port.html

<!--$Id: port.so,v 10.5 2006/08/24 16:32:28 bostic Exp $-->
<!--Copyright (c) 1997,2008 Oracle.  All rights reserved.-->
<!--See the file LICENSE for redistribution information.-->
<html>
<head>
<title>Berkeley DB Reference Guide: Porting Berkeley DB to new architectures</title>
<meta name="description" content="Berkeley DB: An embedded database programmatic toolkit.">
<meta name="keywords" content="embedded,database,programmatic,toolkit,btree,hash,hashing,transaction,transactions,locking,logging,access method,access methods,Java,C,C++">
</head>
<body bgcolor=white>
<table width="100%"><tr valign=top>
<td><b><dl><dt>Berkeley DB Reference Guide:<dd>Distribution</dl></b></td>
<td align=right><a href="../test/faq.html"><img src="../../images/prev.gif" alt="Prev"></a><a href="../toc.html"><img src="../../images/ref.gif" alt="Ref"></a><a href="../distrib/layout.html"><img src="../../images/next.gif" alt="Next"></a>
</td></tr></table>
<p align=center><b>Porting Berkeley DB to new architectures</b></p>
<p>Berkeley DB is generally easy to port to new architectures.  Berkeley DB was
designed to be as portable as possible, and has been ported to a wide
variety of systems, from Wind River's Tornado system, to VMS, to
Windows/NT and Windows/95, and most existing UNIX platforms.  It runs
on 16, 32 and 64-bit machines, little or big-endian.  The difficulty of
a port depends on how much of the ANSI C and POSIX 1003.1 standards the
new architecture offers.</p>
<p>An abstraction layer separates the main Berkeley DB code from the operating
system and architecture specific components.   This layer is comprised
of approximately 2500 lines of C language code, found in the <b>os</b>
subdirectory of the Berkeley DB distribution.  The following list of files
include functionality that may need to be modified or implemented in
order to support a new architecture.  Within each file, there is usually
one, but sometimes several functions (for example, the
<b>os_alloc.c</b> file contains the malloc, calloc, realloc, free,
and strdup functions).</p>
<table border=1 align=center>
<tr><th>Source file</th><th>Description</th></tr>
<tr><td>os_abs.c</td><td>Return if a filename is an absolute pathname</td></tr>
<tr><td>os_alloc.c</td><td>ANSI C malloc, calloc, realloc, strdup, free front-ends</td></tr>
<tr><td>os_clock.c</td><td>Return the current time-of-day</td></tr>
<tr><td>os_config.c</td><td>Return run-time configuration information</td></tr>
<tr><td>os_dir.c</td><td>Read the filenames from a directory</td></tr>
<tr><td>os_errno.c</td><td>Set/get the ANSI C errno value</td></tr>
<tr><td>os_fid.c</td><td>Create a unique ID for a file</td></tr>
<tr><td>os_fsync.c</td><td>POSIX 1003.1 fsync front-end</td></tr>
<tr><td>os_handle.c</td><td>Open file handles</td></tr>
<tr><td>os_id.c</td><td>Return thread ID</td></tr>
<tr><td>os_map.c</td><td>Map a shared memory area</td></tr>
<tr><td>os_method.c</td><td>Run-time replacement of system calls</td></tr>
<tr><td>os_oflags.c</td><td>Convert POSIX 1003.1 open flags, modes to Berkeley DB flags</td></tr>
<tr><td>os_open.c</td><td>Open file handles</td></tr>
<tr><td>os_region.c</td><td>Map a shared memory area</td></tr>
<tr><td>os_rename.c</td><td>POSIX 1003.1 rename call</td></tr>
<tr><td>os_root.c</td><td>Return if application has special permissions</td></tr>
<tr><td>os_rpath.c</td><td>Return last pathname separator</td></tr>
<tr><td>os_rw.c</td><td>POSIX 1003.1 read/write calls</td></tr>
<tr><td>os_seek.c</td><td>POSIX 1003.1 seek call</td></tr>
<tr><td>os_sleep.c</td><td>Cause a thread of control to release the CPU</td></tr>
<tr><td>os_spin.c</td><td>Return the times to spin while waiting for a mutex</td></tr>
<tr><td>os_stat.c</td><td>POSIX 1003.1 stat call</td></tr>
<tr><td>os_tmpdir.c</td><td>Set the path for temporary files</td></tr>
<tr><td>os_unlink.c</td><td>POSIX 1003.1 unlink call</td></tr>
</table>
<p>All but a few of these files contain relatively trivial pieces of code.
Typically, there is only a single version of the code for all platforms
Berkeley DB supports, and that code lives in the <b>os</b> directory of the
distribution.  Where different code is required, the code is either
conditionally compiled or an entirely different version is written. For
example, VxWorks versions of some of these files can be found in the
distribution directory os_vxworks, and Windows versions can be found in
os_windows.</p>
<p>Historically, there are only two difficult questions to answer for each
new port.  The first question is how to handle shared memory.  In order
to write multiprocess database applications (not multithreaded, but
threads of control running in different address spaces), Berkeley DB must be
able to name pieces of shared memory and access them from multiple
processes.  On UNIX/POSIX systems, we use <b>mmap</b> and
<b>shmget</b> for that purpose, but any interface that provides access
to named shared memory is sufficient.  If you have a simple, flat
address space, you should be able to use the code in
<b>os_vxworks/os_map.c</b> as a starting point for the port.  If you
are not intending to write multiprocess database applications, then
this won't be necessary, as Berkeley DB can simply allocate memory from the
heap if all threads of control will live in a single address space.</p>
<p>The second question is mutex support.  Berkeley DB requires some form of
<b>self-blocking</b> mutual exclusion mutex.  Blocking mutexes are
preferred as they tend to be less CPU-expensive and less likely to cause
thrashing.  If blocking mutexes are not available, however, test-and-set
will work as well.  The code for mutexes is in two places in the system:
the include file <b>dbinc/mutex.h</b>, and the distribution directory
<b>mutex</b>.</p>
<p>Berkeley DB uses the GNU autoconf tools for configuration on almost all of
the platforms it supports.  Specifically, the include file
<b>db_config.h</b> configures the Berkeley DB build.  The simplest way to
begin a port is to configure and build Berkeley DB on a UNIX or UNIX-like
system, and then take the <b>Makefile</b> and <b>db_config.h</b>
file created by that configuration, and modify it by hand to reflect
the needs of the new architecture.  Unless you're already familiar with
the GNU autoconf toolset, we don't recommend you take the time to
integrate your changes back into the Berkeley DB autoconfiguration framework.
Instead, send us context diffs of your changes and any new source files
you created, and we'll integrate the changes into our source tree.</p>
<p>Finally, we're happy to work with you on the port, or potentially, do
the port ourselves, if that is of interest to you.  Regardless, if you
have any porting questions, just let us know, and we will be happy to
answer them.</p>
<table width="100%"><tr><td><br></td><td align=right><a href="../test/faq.html"><img src="../../images/prev.gif" alt="Prev"></a><a href="../toc.html"><img src="../../images/ref.gif" alt="Ref"></a><a href="../distrib/layout.html"><img src="../../images/next.gif" alt="Next"></a>
</td></tr></table>
<p><font size=1>Copyright (c) 1996,2008 Oracle.  All rights reserved.</font>
</body>
</html>