<html xmlns="http://www.w3.org/1999/xhtml" xml:lang="en" lang="en">
<head>
<meta http-equiv="Content-Type" content="text/html; charset=utf-8" />
-<meta name="generator" content="Docutils 0.4: http://docutils.sourceforge.net/" />
+<meta name="generator" content="Docutils 0.3.6: http://docutils.sourceforge.net/" />
<title>Permanent Data Storage (Flash)</title>
<meta name="author" content="David Gay, Jonathan Hui" />
<style type="text/css">
</style>
</head>
<body>
-<div class="document" id="permanent-data-storage-flash">
<h1 class="title">Permanent Data Storage (Flash)</h1>
<table class="docinfo" frame="void" rules="none">
<col class="docinfo-name" />
<tr class="field"><th class="docinfo-name">Type:</th><td class="field-body">Documentary</td>
</tr>
<tr><th class="docinfo-name">Status:</th>
-<td>Draft</td></tr>
+<td>Final</td></tr>
<tr class="field"><th class="docinfo-name">TinyOS-Version:</th><td class="field-body">2.x</td>
</tr>
<tr><th class="docinfo-name">Author:</th>
<td>David Gay, Jonathan Hui</td></tr>
-<tr class="field"><th class="docinfo-name">Draft-Created:</th><td class="field-body">27-Sep-2004</td>
-</tr>
-<tr class="field"><th class="docinfo-name">Draft-Version:</th><td class="field-body">1.2</td>
-</tr>
-<tr class="field"><th class="docinfo-name">Draft-Modified:</th><td class="field-body">2006-07-12</td>
-</tr>
-<tr class="field"><th class="docinfo-name">Draft-Discuss:</th><td class="field-body">TinyOS Developer List <tinyos-devel at mail.millennium.berkeley.edu></td>
-</tr>
</tbody>
</table>
+<div class="document" id="permanent-data-storage-flash">
<div class="note">
<p class="first admonition-title">Note</p>
<p class="last">This memo documents a part of TinyOS for the TinyOS Community, and
of this memo is unlimited. This memo is in full compliance with
TEP 1.</p>
</div>
-<div class="section">
-<h1><a id="abstract" name="abstract">Abstract</a></h1>
+<div class="section" id="abstract">
+<h1><a name="abstract">Abstract</a></h1>
<p>This memo documents a set of hardware-independent interfaces to non-volatile
storage for TinyOS 2.x. It describes some design principles for the HPL and
HAL layers of various flash chips.</p>
</div>
-<div class="section">
-<h1><a id="introduction" name="introduction">1. Introduction</a></h1>
-<p>Flash chips are a form of EEPROM (electrically-eraseable, programmable
+<div class="section" id="introduction">
+<h1><a name="introduction">1. Introduction</a></h1>
+<p>Flash chips are a form of EEPROM (electrically-erasable, programmable
read-only memory), distinguished by a fast erase capability. However,
erases can only be done in large units (from 256B to 128kB depending on the
flash chip). Erases are the only way to switch bits from 0 to 1, and
flash chips by their underlying technology (NOR vs NAND). We also
include a column for Atmel's AT45DB flash chip family, as it has
significantly different tradeoffs than other flash chips:</p>
-<pre class="literal-block">
- NOR AT45DB NAND
-
-Erase : Slow (seconds) Fast (ms) Fast (ms)
-Erase unit : Large (64KB-128KB) Small (256B) Medium (8K-32KB)
-Writes : Slow (100s kB/s) Slow (60kB/s) Fast (MBs/s)
-Write unit : 1 bit 256B 100's of bytes
-Bit-errors : Low Low High (requires ECC,
- bad-block mapping)
-Read : Fast* Slow+I/O bus Fast (but limited by
- I/O bus)
-Erase cycles : 10^4 - 10^5 10^4 ** 10^5 - 10^7
-Intended use : Code storage Data storage Data storage
-Energy/byte : 1uJ 1uJ .01uJ
-
-* Intel Mote2 NOR flash is memory mapped (reads are very fast and can
- directly execute code)
-** Or infinite? Data sheet just says that every page within a sector
- must be written every 10^4 writes within that sector
-</pre>
+<blockquote>
+<table border="1" class="docutils">
+<colgroup>
+<col width="19%" />
+<col width="27%" />
+<col width="25%" />
+<col width="28%" />
+</colgroup>
+<thead valign="bottom">
+<tr><th>X</th>
+<th>NOR
+(ex: ST M25P40,
+Intel PXA27x)</th>
+<th>AT45DB</th>
+<th><dl class="first last docutils">
+<dt>NAND</dt>
+<dd>(ex: Samsung
+K9K1G08R0B)</dd>
+</dl>
+</th>
+</tr>
+</thead>
+<tbody valign="top">
+<tr><td>Erase</td>
+<td>Slow (seconds)</td>
+<td>Fast (ms)</td>
+<td>Fast (ms)</td>
+</tr>
+<tr><td>Erase unit</td>
+<td>Large (64KB-128KB)</td>
+<td>Small (256B)</td>
+<td>Medium (8KB-32KB)</td>
+</tr>
+<tr><td>Writes</td>
+<td>Slow (100s kB/s)</td>
+<td>Slow (60kB/s)</td>
+<td>Fast (MBs/s)</td>
+</tr>
+<tr><td>Write unit</td>
+<td>1 bit</td>
+<td>256B</td>
+<td>100's of bytes</td>
+</tr>
+<tr><td>Bit-errors</td>
+<td>Low</td>
+<td>Low</td>
+<td>High (requires ECC,
+bad-block mapping)</td>
+</tr>
+<tr><td>Read</td>
+<td>Bus limited <a class="footnote-reference" href="#id3" id="id1" name="id1">[*]</a></td>
+<td>Slow+Bus limited</td>
+<td>Bus limited</td>
+</tr>
+<tr><td>Erase cycles</td>
+<td>10^4 - 10^5</td>
+<td>10^4 <a class="footnote-reference" href="#id4" id="id2" name="id2">[†]</a></td>
+<td>10^5 - 10^7</td>
+</tr>
+<tr><td>Intended use</td>
+<td>Code storage</td>
+<td>Data storage</td>
+<td>Data storage</td>
+</tr>
+<tr><td>Energy/byte</td>
+<td>1uJ</td>
+<td>1uJ</td>
+<td>.01uJ</td>
+</tr>
+</tbody>
+</table>
+</blockquote>
+<table class="docutils footnote" frame="void" id="id3" rules="none">
+<colgroup><col class="label" /><col /></colgroup>
+<tbody valign="top">
+<tr><td class="label"><a class="fn-backref" href="#id1" name="id3">[*]</a></td><td>M25P40 reads are limited by the use of a 25MHz SPI bus. The PXA27x flash
+is memory mapped (reads are very fast and can directly execute code).</td></tr>
+</tbody>
+</table>
+<table class="docutils footnote" frame="void" id="id4" rules="none">
+<colgroup><col class="label" /><col /></colgroup>
+<tbody valign="top">
+<tr><td class="label"><a class="fn-backref" href="#id2" name="id4">[†]</a></td><td>Or infinite? Data sheet just says that every page within a sector
+must be written every 10^4 writes within that sector</td></tr>
+</tbody>
+</table>
<p>The energy/byte is the per-byte cost of erasing plus programming. It is
derived from the timing and power consumption of erase and write operations
(for NOR flash, values are for the STMicroelectronics M25P family, for NAND
-flash, values are from a Samsung datasheet). Energy/byte for reads appears
+flash, values are from the Samsung datasheet). Energy/byte for reads appears
to depend mostly on how long the read takes (the power consumptions are
comparable), i.e., on the efficiency of the bus + processor.</p>
<p>Early TinyOS platforms all used a flash chip from the AT45DB
<li>Using a high-level memory-like interface (<tt class="docutils literal"><span class="pre">ByteEEPROMC</span></tt>) with
read, write and logging operations.</li>
<li>Using a simple file system (<tt class="docutils literal"><span class="pre">Matchbox</span></tt>) with sequential-only
-files [<a class="reference" href="#id1">1</a>].</li>
+files [<a class="reference" href="#id5">1</a>].</li>
</ul>
<p>Some more recent platforms use different flash chips: the ST M25P family (Telos
rev. B, eyes) and the Intel Strataflash (Intel Mote2). None of the
reimplemented for these other chips, in part because it does not
support some applications (e.g., network reprogramming) very well.</li>
</ul>
+</div>
+<div class="section" id="storage-in-tinyos-2-x">
+<h1><a name="storage-in-tinyos-2-x">2. Storage in TinyOS 2.x</a></h1>
<p>One approach to hiding the differences between different flash chips is to
provide a disk-like, block interface (with, e.g., 512B blocks). This is the
approach taken by compact flash cards. However, in the context of TinyOS,
requires remapping blocks. We believe that maintaining this remapping
table is too expensive on many mote-class devices.</li>
</ul>
-<p>Another approach to supporting multiple flash chips is to build a
-file system (like Matchbox) which can be implemented for multiple
-flash chips. However, TinyOS is currently targeted at running a
-single application, and many applications know their storage needs
-in advance: for instance, a little space for configuration data, and
-everything else for a log of all sampled data. In such cases, the
-flexibility offered by a filing system (e.g., arbitrary numbers of
-files) is overkill, and may come at the expense of implementation
-and runtime complexity.</p>
-<p>Instead, TinyOS 2.x, divides flash chips into separate volumes (with
-sizes fixed at compile-time). Each volume provides a single storage
-abstraction (the abstraction defines the format). So far there are three
-such abstractions: large objects written in a single session,
-small objects with arbitrary reads and writes, and logs. This approach
-has two advantages:</p>
+<p>A second approach is to provide a generic low-level interface providing
+operations (read, write, erase) corresponding to the basic flash
+operations, along with information describing the flash chip's layout
+(minimum write and erase unit, timing information, etc). However,
+we believe that the large differences between NAND and NOR flash (see the
+table above), in particular the differences in reliability, write and
+erase units, make the design of a useful generic low-level interface
+tricky at best.</p>
+<p>We thus believe it is best, for now at least, to define high-level
+storage abstractions that are useful for sensor network applications,
+and leave their implementation up to each flash chip - such abstractions
+will be necessary anyway. We leave open the possibility that a future
+TEP may define portable lower-level flash interfaces (either for all
+flash chips, or, e.g., for NOR-family flashes). Such low-level
+interfaces would allow implementations of the storage abstractions
+defined in this TEP to be used for multiple flash chips.</p>
+<p>This TEP describes three high-level storage abstractions: large objects
+written in a single session, small objects with arbitrary reads and
+writes, and logs. TinyOS 2.x, divides flash chips into separate volumes
+(with sizes fixed at compile-time). Each volume provides a single
+storage abstraction (the abstraction defines the format).</p>
+<p>We prefer the use of single abstractions over fixed-size volumes over
+the use of a more general filing system (like Matchbox) for several
+reasons:</p>
<ul class="simple">
+<li>TinyOS is currently targeted at running a single application, and many
+applications know their storage needs in advance: for instance, a
+little space for configuration data, and everything else for a log of
+all sampled data. In such cases, the flexibility offered by a filing
+system (e.g., arbitrary numbers of files) is overkill,</li>
<li>Each abstraction is relatively easy to implement on a new flash chip, and
has relatively little overhead.</li>
<li>The problem of dealing with the limited number of erase cycles/block
"wear levelling" (ensuring that each block of the flash is erased
the same number of time, to maximise flash chip lifetime).</li>
</ul>
-<p>New abstractions (including a filing system) can easily be added to this
-framework, or can be built on top of these abstractions.</p>
+<p>New abstractions (including a filing system) can easily be added to
+this framework.</p>
<p>The rest of this TEP covers some principles for the organisation of
-flash chips (Section 2), then describes the flash volumes and
-storage abstractions in detail (Section 3).</p>
+flash chips (Section 3), then describes the flash volumes and
+storage abstractions in detail (Section 4).</p>
</div>
-<div class="section">
-<h1><a id="hpl-hal-hil-architecture" name="hpl-hal-hil-architecture">2. HPL/HAL/HIL Architecture</a></h1>
-<p>The flash chip architecture dollows the three-layer Hardware
+<div class="section" id="hpl-hal-hil-architecture">
+<h1><a name="hpl-hal-hil-architecture">3. HPL/HAL/HIL Architecture</a></h1>
+<p>The flash chip architecture follows the three-layer Hardware
Abstraction Architecture (HAA), with each chip providing a presentation
-layer (HPL, Section 2.1), adaptation layer (HAL, Section 2.2) and
-platform-independent interface layer (the storage abstractions described in
-Section 3) [<a class="reference" href="#id2">2</a>]. The implementation of these layers SHOULD be found in the
+layer (HPL, Section 3.1), adaptation layer (HAL, Section 3.2) and
+platform-independent interface layer (HIL, Section 3.3) [<a class="reference" href="#id6">2</a>].
+The implementation of these layers SHOULD be found in the
<tt class="docutils literal"><span class="pre">tos/chips/CHIPNAME</span></tt> directory. If a flash chip is part of a larger
family with a similar interface, the HAA SHOULD support all family members
by relying, e.g., on platform-provided configuration information.</p>
<p>Appendix A shows example HPL and HAL specifications for the AT45DB
and ST M25P chip families.</p>
-<div class="section">
-<h2><a id="hardware-presentation-layer-hpl" name="hardware-presentation-layer-hpl">2.1 Hardware Presentation Layer (HPL)</a></h2>
-<p>The flash HPL has a chip-dependent, system-independent interface. The
-implementation of this HPL is system-dependent. The flash HPL SHOULD be
+<div class="section" id="hardware-presentation-layer-hpl">
+<h2><a name="hardware-presentation-layer-hpl">3.1 Hardware Presentation Layer (HPL)</a></h2>
+<p>The flash HPL has a chip-dependent, platform-independent interface. The
+implementation of this HPL is platform-dependent. The flash HPL SHOULD be
stateless.</p>
<p>To remain platform independent, a flash chip's HPL SHOULD connect to
platform-specific components
flash chips, this directory MAY also contain a file describing the
particular flash chip found on the platform.</p>
</div>
-<div class="section">
-<h2><a id="hardware-adaptation-layer-hal" name="hardware-adaptation-layer-hal">2.2 Hardware Adaptation Layer (HAL)</a></h2>
-<p>The flash HAL has a chip-dependent, system-independent interface and
+<div class="section" id="hardware-adaptation-layer-hal">
+<h2><a name="hardware-adaptation-layer-hal">3.2 Hardware Adaptation Layer (HAL)</a></h2>
+<p>The flash HAL has a chip-dependent, platform-independent interface and
implementation. Flash families with a common HPL SHOULD have a common
HAL. Flash HAL's SHOULD expose a <tt class="docutils literal"><span class="pre">Resource</span></tt> interface and automatically
power-manage the underlying flash chip. Finally, the flash HAL MUST
programmer (see Section 3). This allows users to build new flash
abstractions that interact cleanly with the rest of the flash system.</p>
</div>
+<div class="section" id="hardware-interface-layer-hil">
+<h2><a name="hardware-interface-layer-hil">3.3 Hardware Interface Layer (HIL)</a></h2>
+<p>Each flash chip MUST support at least one of the storage abstractions
+described in Section 4. These abstractions SHOULD be presented in
+components named <tt class="docutils literal"><span class="pre">ChipAbstractionC</span></tt>, e.g., <tt class="docutils literal"><span class="pre">At45dbLogStorageC</span></tt>.
+Additionally, a flash chip implementation MAY support platforms
+with multiple instances of the same storage chip. The way in which
+this is achieved is not specified further in this TEP.</p>
+<p>Each platform MUST have <tt class="docutils literal"><span class="pre">AbstractionC</span></tt> components (e.g.,
+<tt class="docutils literal"><span class="pre">LogStorageC</span></tt>) implementing the storage abstractions of Section 4
+supported by its flash chip(s). On platforms with multiple storage chips
+SHOULD redirect uses of <tt class="docutils literal"><span class="pre">AbstractionC</span></tt> to the appropriate storage
+chip, based on the requested volume.</p>
</div>
-<div class="section">
-<h1><a id="non-volatile-storage-abstracitons-in-tinyos-2-x" name="non-volatile-storage-abstracitons-in-tinyos-2-x">3. Non-Volatile Storage Abstracitons in TinyOS 2.x</a></h1>
-<p>The HIL implementations are system-independent, but chip (family)
+</div>
+<div class="section" id="non-volatile-storage-abstractions">
+<h1><a name="non-volatile-storage-abstractions">4. Non-Volatile Storage Abstractions</a></h1>
+<p>The HIL implementations are platform-independent, but chip (family)
dependent. They implement the three storage abstractions and
volume structure discussed in the introduction.</p>
-<div class="section">
-<h2><a id="volumes" name="volumes">3.1. Volumes</a></h2>
+<div class="section" id="volumes">
+<h2><a name="volumes">4.1. Volumes</a></h2>
<p>The division of the flash chip into fixed-size volumes is specified by
an XML file that is placed in the application's directory (where one
types 'make'). The XML file specifies the allocation as follows:</p>
except that the specification to physical allocation MUST be one-to-one
(i.e. a given specification should always have the same resulting physical
allocation on a given chip) and the result MUST be placed in the build
-directory. When not specified, the tool may give any suitable physical
-location to a volume. If there is any reason that the physical allocation
+directory. When not specified, the tool picks a suitable physical
+location for a volume. If there is any reason that the physical allocation
cannot be satisfied, an error should be given at compile time. The tool
SHOULD be named <tt class="docutils literal"><span class="pre">tos-storage-CHIPNAME</span></tt> and be distributed with the other
-tools supporting a platform.</p>
+tools supporting a platform. The XML file SHOULD be named
+<tt class="docutils literal"><span class="pre">volumes-CHIPNAME.xml</span></tt>.</p>
<p>The compile-time tool MUST prepend 'VOLUME_' to each volume name in
the XML file and '#define' each resulting name to map to a unique
integer.</p>
compile-time error since the symbol will be undefined.</p>
<p>A volume MUST NOT be used with more than one storage abstraction instance.</p>
</div>
-<div class="section">
-<h2><a id="large-objects" name="large-objects">3.2 Large objects</a></h2>
-<p>The motivating example for large objects is the transmission or long-term
-storage of large pieces of data. For instance, programs in a network-reprogramming
-system, or large data-packets in a reliable data-transmission system. Such
-objects have two interesting characteristics: each byte in the object is
-written at most once, and a full object is written in a single "session"
-(i.e., without the mote rebooting).</p>
+<div class="section" id="large-objects">
+<h2><a name="large-objects">4.2 Large objects</a></h2>
+<p>The motivating example for large objects is the transmission or
+long-term storage of large pieces of data. For instance, programs in a
+network-reprogramming system, or large data-packets in a reliable
+data-transmission system. Such objects have an interesting
+characteristic: each byte in the object is written at most once.</p>
<p>This leads to the definition of the <tt class="docutils literal"><span class="pre">BlockStorageC</span></tt> abstraction for storing
-large objects:</p>
+large objects or other "write-once" objects:</p>
<ul class="simple">
<li>A large object ranges from a few kilobytes upwards.</li>
-<li>A large object must be erased before use.</li>
-<li>A large object must be committed to ensure it survives a reboot or crash;
-after a commit no more writes may be performed.</li>
-<li>Random reads are allowed.</li>
-<li>Random writes are allowed are allowed between erase and commit; data
-cannot be overwritten.</li>
+<li>A large object is erased before the first write.</li>
+<li>A sync ensures that a large object survives a reboot or crash</li>
+<li>Reads are unrestricted</li>
+<li>Each byte can only be written once between two erases</li>
</ul>
<p>Large objects are accessed by instantiating a BlockStorageC component
which takes a volume id argument:</p>
}
} ...
</pre>
-<p>The <tt class="docutils literal"><span class="pre">BlockRead</span></tt> and <tt class="docutils literal"><span class="pre">BlockWrite</span></tt> interfaces contain the following
-operations (all split-phase, except <tt class="docutils literal"><span class="pre">BlockRead.getSize</span></tt>):</p>
+<p>The <tt class="docutils literal"><span class="pre">BlockRead</span></tt> and <tt class="docutils literal"><span class="pre">BlockWrite</span></tt> interfaces (briefly presented in
+Appendix B) contain the following operations (all split-phase, except
+<tt class="docutils literal"><span class="pre">BlockRead.getSize</span></tt>):</p>
<ul class="simple">
<li><tt class="docutils literal"><span class="pre">BlockWrite.erase</span></tt>: erase the volume. After a reboot or a commit, a
-volume must be erased before it can be written to.</li>
-<li><tt class="docutils literal"><span class="pre">BlockWrite.write</span></tt>: write some bytes starting at a given offset. Each
-byte can only be written once between an erase and the subsequent commit.</li>
-<li><tt class="docutils literal"><span class="pre">BlockWrite.commit</span></tt>: commit all writes to a given volume. No writes can
-be performed after a commit until a subsequent erase.</li>
-<li><tt class="docutils literal"><span class="pre">BlockRead.verify</span></tt>: verify that the volume contains the results of a
-successful commit.</li>
+volume MUST be erased before it can be written to.</li>
+<li><tt class="docutils literal"><span class="pre">BlockWrite.write</span></tt>: write some bytes starting at a given
+offset. Each byte MUST NOT be written more than once between two erases.</li>
+<li><tt class="docutils literal"><span class="pre">BlockWrite.sync</span></tt>: ensure all previous writes are present on a given
+volume. Sync MUST be called to ensure written data survives a reboot
+or crash.</li>
<li><tt class="docutils literal"><span class="pre">BlockRead.read</span></tt>: read some bytes starting at a given offset.</li>
<li><tt class="docutils literal"><span class="pre">BlockRead.computeCrc</span></tt>: compute the CRC of some bytes starting at a
given offset.</li>
</ul>
<p>For full details on arguments and other considerations, see the comments in
the interface definitions.</p>
+<p>Note that these interfaces contain no direct support for verifying the
+integrity of the BlockStorage data, but such support can easily be built
+by using the <tt class="docutils literal"><span class="pre">computeCrc</span></tt> command and storing the result in a
+well-defined location, and checking this CRC when desired.</p>
</div>
-<div class="section">
-<h2><a id="logging" name="logging">3.3 Logging</a></h2>
-<p>Event and reuslt logging is a common requirement in sensor
+<div class="section" id="logging">
+<h2><a name="logging">4.3 Logging</a></h2>
+<p>Event and result logging is a common requirement in sensor
networks. Such logging should be reliable (a mote crash should not
lose data). It should also be easy to extract data from the log,
either partially or fully. Some logs are <em>linear</em> (stop logging when
the volume is full), others are <em>circular</em> (the oldest data is
overwritten when the volume is full).</p>
-<p>The <tt class="docutils literal"><span class="pre">LogStorageC</span></tt> abstraction supports these requirements. The log is record
-based: each call to <tt class="docutils literal"><span class="pre">LogWrite.append</span></tt> (see below) creates a new
-record. On failure (crash or reboot), the log is guaranteed to only lose
-whole records from the end of the log. Additionally, once a circular log
-wraps around, calls to <tt class="docutils literal"><span class="pre">LogWrite.append</span></tt> only lose whole records from the
-beginning of the log. These guarantees mean that applications do not to
-have worry about incomplete or inconsistent log entries.</p>
+<p>The <tt class="docutils literal"><span class="pre">LogStorageC</span></tt> abstraction supports these requirements. The log is
+record based: each call to <tt class="docutils literal"><span class="pre">LogWrite.append</span></tt> (see below) creates a new
+record. On failure (crash or reboot), the log MUST only lose whole
+records from the end of the log. Additionally, once a circular log wraps
+around, calls to <tt class="docutils literal"><span class="pre">LogWrite.append</span></tt> MUST only lose whole records from
+the beginning of the log.</p>
<p>Logs are accessed by instantiating a LogStorageC component which takes a
volume id and a boolean argument:</p>
<pre class="literal-block">
</pre>
<p>If the <tt class="docutils literal"><span class="pre">circular</span></tt> argument is TRUE, the log is circular; otherwise
it is linear.</p>
-<p>The <tt class="docutils literal"><span class="pre">LogRead</span></tt> and <tt class="docutils literal"><span class="pre">LogWrite</span></tt> interfaces contain the following
-operations (all split-phase except <tt class="docutils literal"><span class="pre">LogWrite.currentOffset</span></tt>,
-<tt class="docutils literal"><span class="pre">LogRead.currentOffset</span></tt> and <tt class="docutils literal"><span class="pre">LogRead.getSize</span></tt>):</p>
+<p>The <tt class="docutils literal"><span class="pre">LogRead</span></tt> and <tt class="docutils literal"><span class="pre">LogWrite</span></tt> interfaces (briefly presented in
+Appendix B) contain the following operations (all split-phase except
+<tt class="docutils literal"><span class="pre">LogWrite.currentOffset</span></tt>, <tt class="docutils literal"><span class="pre">LogRead.currentOffset</span></tt> and
+<tt class="docutils literal"><span class="pre">LogRead.getSize</span></tt>):</p>
<ul>
-<li><p class="first"><tt class="docutils literal"><span class="pre">LogWrite.erase</span></tt>: erase the log.</p>
+<li><p class="first"><tt class="docutils literal"><span class="pre">LogWrite.erase</span></tt>: erase the log. A log MUST be erased (possibly in
+some previous session) before any other operation can be used.</p>
</li>
<li><p class="first"><tt class="docutils literal"><span class="pre">LogWrite.append</span></tt>: append some bytes to the log. In a circular log,
this may overwrite the current read position. In this case, the
-read position is implicitly advanced to the log's current beginning
-(i.e., as if <tt class="docutils literal"><span class="pre">LogRead.seek</span></tt> had been called with <tt class="docutils literal"><span class="pre">SEEK_BEGINNING</span></tt>).</p>
+read position MUST be advanced to the log's current beginning
+(i.e., as if <tt class="docutils literal"><span class="pre">LogRead.seek</span></tt> had been called with <tt class="docutils literal"><span class="pre">SEEK_BEGINNING</span></tt>).
+Additionally, the <tt class="docutils literal"><span class="pre">LogWrite.appendDone</span></tt> event reports whenever, in a
+circular log, an append MAY have erased old records.</p>
<p>Each append creates a separate record. Log implementations may have a
maximum record size; all implementations MUST support records of up
to 255 bytes.</p>
</li>
<li><p class="first"><tt class="docutils literal"><span class="pre">LogWrite.sync</span></tt>: guarantee that data written so far will not be lost to
a crash or reboot (it can still be overwritten when a circular log wraps
-around). Using <tt class="docutils literal"><span class="pre">sync</span></tt> may waste some space in the log.</p>
+around). Using <tt class="docutils literal"><span class="pre">sync</span></tt> MAY waste some space in the log.</p>
</li>
<li><p class="first"><tt class="docutils literal"><span class="pre">LogWrite.currentOffset</span></tt>: return cookie representing current
append position (for use with <tt class="docutils literal"><span class="pre">LogRead.seek</span></tt>).</p>
</li>
<li><p class="first"><tt class="docutils literal"><span class="pre">LogRead.read</span></tt>: read some bytes from the current read position in
the log and advance the read position.</p>
+<p><tt class="docutils literal"><span class="pre">LogStorageC</span></tt> implementations MUST include error detection codes
+to increase the likelihood of detection of corrupted or invalid log
+data. Data returned by a successful read MUST have passed this
+error detection check. The behaviour on failure of this check is
+unspecified (e.g., the at45db believes as if the end of the log has
+been reached; other implementations may behave differently).</p>
</li>
<li><p class="first"><tt class="docutils literal"><span class="pre">LogRead.currentOffset</span></tt>: return cookie representing current
read position (for use with <tt class="docutils literal"><span class="pre">LogRead.seek</span></tt>).</p>
<tt class="docutils literal"><span class="pre">SEEK_BEGINNING</span></tt> was requested.</p>
<p><tt class="docutils literal"><span class="pre">SEEK_BEGINNING</span></tt> positions the read position at the beginning of
the oldest record still present in the log.</p>
+<p>After reboot, the current read position is <tt class="docutils literal"><span class="pre">SEEK_BEGINNING</span></tt>.</p>
</li>
-<li><p class="first"><tt class="docutils literal"><span class="pre">LogRead.getSize</span></tt>: return an approximation of the log's capacity.
-Uses of <tt class="docutils literal"><span class="pre">sync</span></tt> and other overhead may reduce this number.</p>
+<li><p class="first"><tt class="docutils literal"><span class="pre">LogRead.getSize</span></tt>: return an approximation of the log's capacity
+in bytes. Uses of <tt class="docutils literal"><span class="pre">sync</span></tt> and other overhead may reduce this number.</p>
</li>
</ul>
<p>For full details on arguments, etc, see the comments in the interface
definitions.</p>
+<p>Note that while each call to <tt class="docutils literal"><span class="pre">append</span></tt> logically creates a separate
+record, the <tt class="docutils literal"><span class="pre">LogStorageC</span></tt> API does not report record
+boundaries. Additionally, crashes, reboots, and appends after
+wrap-around in a circular log can cause the loss of multiple consecutive
+records. Taken together, these restrictions mean that a <tt class="docutils literal"><span class="pre">LogStorageC</span></tt>
+implementation MAY internally aggregate several consecutive appends into
+a single record. However, the guarantee that only whole records are lost
+is sufficient to ensure that applications do not to have worry about
+incomplete or inconsistent log entries.</p>
</div>
-<div class="section">
-<h2><a id="small-objects" name="small-objects">3.4 Small objects:</a></h2>
-<p>Sensor network applications may need to store configuration data, e.g.,
-mote id, radio frequency, sample rates, etc. Such data is not large, but
+<div class="section" id="small-objects">
+<h2><a name="small-objects">4.4 Small objects:</a></h2>
+<p>Sensor network applications need to store configuration data, e.g.,
+mote identity, radio frequency, sample rates, etc. Such data is not large, but
losing it may lead to a mote misbehaving or losing contact with the
network.</p>
<p>The <tt class="docutils literal"><span class="pre">ConfigStorageC</span></tt> abstraction stores a single small object in a volume. It:</p>
</pre>
<p>A small object MUST be mounted (via the <tt class="docutils literal"><span class="pre">Mount</span></tt> interface) before
the first use.</p>
-<p>The <tt class="docutils literal"><span class="pre">Mount</span></tt> and <tt class="docutils literal"><span class="pre">ConfigStorage</span></tt> interfaces contain the following
-operations (all split-phase except <tt class="docutils literal"><span class="pre">ConfigStorage.getSize</span></tt> and
-<tt class="docutils literal"><span class="pre">ConfigStorage.valid</span></tt>):</p>
+<p>The <tt class="docutils literal"><span class="pre">Mount</span></tt> and <tt class="docutils literal"><span class="pre">ConfigStorage</span></tt> interfaces (briefly presented in
+Appendix B) contain the following operations (all split-phase except
+<tt class="docutils literal"><span class="pre">ConfigStorage.getSize</span></tt> and <tt class="docutils literal"><span class="pre">ConfigStorage.valid</span></tt>):</p>
<ul class="simple">
<li><tt class="docutils literal"><span class="pre">Mount.mount</span></tt>: mount the volume.</li>
<li><tt class="docutils literal"><span class="pre">ConfigStorage.valid</span></tt>: return TRUE if the volume contains a
definitions.</p>
</div>
</div>
-<div class="section">
-<h1><a id="implementations" name="implementations">4. Implementations</a></h1>
+<div class="section" id="implementations">
+<h1><a name="implementations">5. Implementations</a></h1>
<p>An AT45DB implementation can be found in tinyos-2.x/tos/chips/at45db.</p>
<p>An ST M25P implementation can be found in tinyos-2.x/tos/chips/stm25p.</p>
</div>
-<div class="section">
-<h1><a id="authors-addresses" name="authors-addresses">5. Authors' Addresses</a></h1>
+<div class="section" id="authors-addresses">
+<h1><a name="authors-addresses">6. Authors' Addresses</a></h1>
<div class="line-block">
<div class="line">David Gay</div>
<div class="line">2150 Shattuck Ave, Suite 1300</div>
<div class="line">email - <a class="reference" href="mailto:jhui@archedrock.com">jhui@archedrock.com</a></div>
</div>
</div>
-<div class="section">
-<h1><a id="citations" name="citations">6. Citations</a></h1>
-<table class="docutils footnote" frame="void" id="id1" rules="none">
+<div class="section" id="citations">
+<h1><a name="citations">7. Citations</a></h1>
+<table class="docutils footnote" frame="void" id="id5" rules="none">
<colgroup><col class="label" /><col /></colgroup>
<tbody valign="top">
-<tr><td class="label"><a name="id1">[1]</a></td><td>David Gay. "Design of Matchbox, the simple filing system for
+<tr><td class="label"><a name="id5">[1]</a></td><td>David Gay. "Design of Matchbox, the simple filing system for
motes. (version 1.0)."</td></tr>
</tbody>
</table>
-<table class="docutils footnote" frame="void" id="id2" rules="none">
+<table class="docutils footnote" frame="void" id="id6" rules="none">
<colgroup><col class="label" /><col /></colgroup>
<tbody valign="top">
-<tr><td class="label"><a name="id2">[2]</a></td><td>TEP 2: Hardware Abstraction Architecture.</td></tr>
+<tr><td class="label"><a name="id6">[2]</a></td><td>TEP 2: Hardware Abstraction Architecture.</td></tr>
</tbody>
</table>
</div>
-<div class="section">
-<h1><a id="appendix-a-haa-for-some-existing-flash-chips" name="appendix-a-haa-for-some-existing-flash-chips">Appendix A. HAA for some existing flash chips</a></h1>
-<div class="section">
-<h2><a id="a-1-at45db" name="a-1-at45db">A.1 AT45DB</a></h2>
+<div class="section" id="appendix-a-haa-for-some-existing-flash-chips">
+<h1><a name="appendix-a-haa-for-some-existing-flash-chips">Appendix A. HAA for some existing flash chips</a></h1>
+<div class="section" id="a-1-at45db">
+<h2><a name="a-1-at45db">A.1 AT45DB</a></h2>
<p>The Atmel AT45DB family HPL is:</p>
<pre class="literal-block">
configuration HplAt45dbC {
<p>The <tt class="docutils literal"><span class="pre">At45dbVolume</span></tt> interface has operations to report volume size and
map volume-relative pages to absolute pages.</p>
</div>
-<div class="section">
-<h2><a id="a-2-st-m25p" name="a-2-st-m25p">A.2 ST M25P</a></h2>
+<div class="section" id="a-2-st-m25p">
+<h2><a name="a-2-st-m25p">A.2 ST M25P</a></h2>
<p>The ST M25P family HPL is:</p>
<pre class="literal-block">
configuration Stm25pSpiC {
obtain the volume id of each of its clients.</p>
</div>
</div>
+<div class="section" id="appendix-b-storage-interfaces">
+<h1><a name="appendix-b-storage-interfaces">Appendix B. Storage interfaces</a></h1>
+<p>These interfaces are presented briefly here for reference; please refer
+to the TinyOS documentation for a full description of the commands,
+events and their parameters.</p>
+<div class="section" id="b-1-blockstorage-interfaces">
+<h2><a name="b-1-blockstorage-interfaces">B.1 BlockStorage interfaces</a></h2>
+<p>The BlockStorage interfaces are:</p>
+<pre class="literal-block">
+interface BlockRead {
+ command error_t read(storage_addr_t addr, void* buf, storage_len_t len);
+ event void readDone(storage_addr_t addr, void* buf, storage_len_t len,
+ error_t error);
+
+ command error_t computeCrc(storage_addr_t addr, storage_len_t len,
+ uint16_t crc);
+ event void computeCrcDone(storage_addr_t addr, storage_len_t len,
+ uint16_t crc, error_t error);
+
+ command storage_len_t getSize();
+}
+
+interface BlockWrite {
+ command error_t write(storage_addr_t addr, void* buf, storage_len_t len);
+ event void writeDone(storage_addr_t addr, void* buf, storage_len_t len,
+ error_t error);
+
+ command error_t erase();
+ event void eraseDone(error_t error);
+
+ command error_t sync();
+ event void syncDone(error_t error);
+}
+</pre>
+</div>
+<div class="section" id="b-2-configstorage-interfaces">
+<h2><a name="b-2-configstorage-interfaces">B.2 ConfigStorage interfaces</a></h2>
+<p>The ConfigStorage interfaces are:</p>
+<pre class="literal-block">
+interface Mount {
+ command error_t mount();
+ event void mountDone(error_t error);
+}
+
+interface ConfigStorage {
+ command error_t read(storage_addr_t addr, void* buf, storage_len_t len);
+ event void readDone(storage_addr_t addr, void* buf, storage_len_t len,
+ error_t error);
+
+ command error_t write(storage_addr_t addr, void* buf, storage_len_t len);
+ event void writeDone(storage_addr_t addr, void* buf, storage_len_t len,
+ error_t error);
+
+ command error_t commit();
+ event void commitDone(error_t error);
+
+ command storage_len_t getSize();
+ command bool valid();
+}
+</pre>
+</div>
+<div class="section" id="b-3-logstorage-interfaces">
+<h2><a name="b-3-logstorage-interfaces">B.3 LogStorage interfaces</a></h2>
+<p>The LogStorage interfaces are:</p>
+<pre class="literal-block">
+interface LogRead {
+ command error_t read(void* buf, storage_len_t len);
+ event void readDone(void* buf, storage_len_t len, error_t error);
+
+ command storage_cookie_t currentOffset();
+
+ command error_t seek(storage_cookie_t offset);
+ event void seekDone(error_t error);
+
+ command storage_len_t getSize();
+}
+
+interface LogWrite {
+ command error_t append(void* buf, storage_len_t len);
+ event void appendDone(void* buf, storage_len_t len, bool recordsLost,
+ error_t error);
+
+ command storage_cookie_t currentOffset();
+
+ command error_t erase();
+ event void eraseDone(error_t error);
+
+ command error_t sync();
+ event void syncDone(error_t error);
+}
+</pre>
+</div>
+</div>
</div>
</body>
</html>
projects / tinyos-2.x.git / blobdiff