Gitiles
Code Review Sign In
gerrit-public.fairphone.software / platform / external / llvm / b39021b8ba9d2450f3f5a6d9aec42e955d0acb2b / . / docs / BytecodeFormat.html
blob: 287efd13213f7bf38f34c0118abaea5ec8f5347b [file] [log] [blame]
Reid Spencer50026612004-05-22 02:28:36 +0000[diff] [blame]1<!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01//EN"
2 "http://www.w3.org/TR/html4/strict.dtd">
3<html>
4<head>
5 <title>LLVM Bytecode File Format</title>
6 <link rel="stylesheet" href="llvm.css" type="text/css">
7 <style>
8 table, tr, td { border: 2px solid gray }
Reid Spencerb39021b2004-05-23 17:05:09 +0000[diff] [blame^]9 th { border: 2px solid gray; font-weight: bold; }
Reid Spencer50026612004-05-22 02:28:36 +0000[diff] [blame]10 table { border-collapse: collapse; margin-top: 1em margin-bottom: 1em }
11 </style>
12</head>
13<body>
14 <div class="doc_title"> LLVM Bytecode File Format </div>
15<ol>
16 <li><a href="#abstract">Abstract</a></li>
17 <li><a href="#general">General Concepts</a></li>
18 <ol>
19 <li><a href="#blocks">Blocks</a></li>
20 <li><a href="#lists">Lists</a></li>
21 <li><a href="#fields">Fields</a></li>
Reid Spencerb39021b2004-05-23 17:05:09 +0000[diff] [blame^]22 <li><a href="#encoding">Encoding Rules</a></li>
Reid Spencer50026612004-05-22 02:28:36 +0000[diff] [blame]23 <li><a href="#align">Alignment</a></li>
24 </ol>
25 <li><a href="#details">Detailed Layout</a>
26 <ol>
27 <li><a href="#notation">Notation</a></li>
28 <li><a href="#blocktypes">Blocks Types</a></li>
Reid Spencerb39021b2004-05-23 17:05:09 +0000[diff] [blame^]29 <li><a href="#signature">Signature Block</a></li>
30 <li><a href="#module">Module Block</a></li>
Reid Spencer50026612004-05-22 02:28:36 +0000[diff] [blame]31 <li><a href="#typeool">Global Type Pool</a></li>
32 <li><a href="#modinfo">Module Info Block</a></li>
33 <li><a href="#constants">Global Constant Pool</a></li>
34 <li><a href="#functions">Function Blocks</a><li>
35 <li><a href="#symtab">Module Symbol Table</a><li>
36 </ol>
37 </li>
38</ol>
39<div class="doc_text">
40<p><b>Written by <a href="mailto:rspencer@x10sys.com">Reid Spencer</a>
41and <a href="mailto:sabre@nondot.org">Chris Lattner</a></b></p>
42<p> </p>
43</div>
44<!-- *********************************************************************** -->
45<div class="doc_section"> <a name="abstract">Abstract </a></div>
46<!-- *********************************************************************** -->
47<div class="doc_text">
48<p>This document is an (after the fact) specification of the LLVM bytecode
49file format. It documents the binary encoding rules of the bytecode file format
50so that equivalent systems can encode bytecode files correctly. The LLVM
51bytecode representation is used to store the intermediate representation on
52disk in compacted form.
53</p>
54</div>
55<!-- *********************************************************************** -->
56<div class="doc_section"> <a name="general">General Concepts</a> </div>
57<!-- *********************************************************************** -->
58<div class="doc_text">
59<p>This section describes the general concepts of the bytecode file format
60without getting into bit and byte level specifics.</p>
61</div>
62<!-- _______________________________________________________________________ -->
63<div class="doc_subsection"><a name="blocks">Blocks</a> </div>
64<div class="doc_text">
65<p>LLVM bytecode files consist simply of a sequence of blocks of bytes.
66Each block begins with an identification value that determines the type of
67the next block. The possible types of blocks are described below in the section
Reid Spencerb39021b2004-05-23 17:05:09 +0000[diff] [blame^]68<a href="#blocktypes">Block Types</a>. The block identifier is used because
Reid Spencer50026612004-05-22 02:28:36 +0000[diff] [blame]69it is possible for entire blocks to be omitted from the file if they are
70empty. The block identifier helps the reader determine which kind of block is
71next in the file.</p>
Reid Spencer939290f2004-05-22 05:56:41 +0000[diff] [blame]72<p>The following block identifiers are currently in use
73(from llvm/Bytecode/Format.h):</p>
74<ol>
75 <li><b>Module (0x01)</b>.</li>
76 <li><b>Function (0x11)</b>.</li>
77 <li><b>ConstantPool (0x12)</b>.</li>
78 <li><b>SymbolTable (0x13)</b>.</li>
79 <li><b>ModuleGlobalInfo (0x14)</b>.</li>
80 <li><b>GlobalTypePlane (0x15)</b>.</li>
81 <li><b>BasicBlock (0x31)</b>.</li>
82 <li><b>InstructionList (0x32)</b>.</li>
83 <li><b>CompactionTable (0x33)</b>.</li>
84</ol>
Reid Spencerb39021b2004-05-23 17:05:09 +0000[diff] [blame^]85<p> All blocks are variable length. They consume just enough bytes to express
86their contents. Each block begins with an integer identifier and the length
87of the block.</p>
Reid Spencer50026612004-05-22 02:28:36 +0000[diff] [blame]88</div>
89<!-- _______________________________________________________________________ -->
90<div class="doc_subsection"><a name="lists">Lists</a> </div>
91<div class="doc_text">
92<p>Most blocks are constructed of lists of information. Lists can be constructed
93of other lists, etc. This decomposition of information follows the containment
94hierarchy of the LLVM Intermediate Representation. For example, a function is
95composed of a list of basic blocks. Each basic block is composed of a set of
96instructions. This list of list nesting and hierarchy is maintained in the
97bytecode file.</p>
98<p>A list is encoded into the file simply by encoding the number of entries as
99an integer followed by each of the entries. The reader knows when the list is
100done because it will have filled the list with the required numbe of entries.
101</p>
102</div>
103<!-- _______________________________________________________________________ -->
104<div class="doc_subsection"><a name="fields">Fields</a> </div>
105<div class="doc_text">
106<p>Fields are units of information that LLVM knows how to write atomically.
107Most fields have a uniform length or some kind of length indication built into
108their encoding. For example, a constant string (array of SByte or UByte) is
109written simply as the length followed by the characters. Although this is
110similar to a list, constant strings are treated atomically and are thus
111fields.</p>
112<p>Fields use a condensed bit format specific to the type of information
113they must contain. As few bits as possible are written for each field. The
114sections that follow will provide the details on how these fields are
115written and how the bits are to be interpreted.</p>
116</div>
117<!-- _______________________________________________________________________ -->
Reid Spencerb39021b2004-05-23 17:05:09 +0000[diff] [blame^]118<div class="doc_subsection"><a name="encoding">Encoding Primitives</a> </div>
119<div class="doc_text">
120<p>Each field that can be put out is encoded into the file using a small set
121of primitives. The rules for these primitives are described below.</p>
122<h3>Variable Bit Rate Encoding</h3>
123<p>To minimize the number of bytes written for small quantities, an encoding
124scheme similar to UTF-8 is used to write integer data. The scheme is known as
125variable bit rate (vbr) encoding. In this encoding, the high bit of each
126byte is used to indicate if more bytes follow. If (byte &amp; 0x80) is non-zero
127in any given byte, it means there is another byte immediately following that
128also contributes to the value. For the final byte (byte &amp; 0x80) is false
129(the high bit is not set). In each byte only the low seven bits contribute to
130the value. Consequently 32-bit quantities can take from one to <em>five</em>
131bytes to encode. In general, smaller quantities will encode in fewer bytes,
132as follows:</p>
133<table class="doc_table_nw">
134 <tr>
135 <th>Byte #</th>
136 <th>Significant Bits</th>
137 <th>Maximum Value</th>
138 </tr>
139 <tr><td>1</td><td>0-6</td><td>127</td></tr>
140 <tr><td>2</td><td>7-13</td><td>16,383</td></tr>
141 <tr><td>3</td><td>14-20</td><td>2,097,151</td></tr>
142 <tr><td>4</td><td>21-27</td><td>268,435,455</td></tr>
143 <tr><td>5</td><td>28-34</td><td>34,359,738,367</td></tr>
144 <tr><td>6</td><td>35-41</td><td>4,398,046,511,103</td></tr>
145 <tr><td>7</td><td>42-48</td><td>562,949,953,421,311</td></tr>
146 <tr><td>8</td><td>49-55</td><td>72,057,594,037,927,935</td></tr>
147 <tr><td>9</td><td>56-62</td><td>9,223,372,036,854,775,807</td></tr>
148 <tr><td>10</td><td>63-69</td><td>1,180,591,620,717,411,303,423</td></tr>
149</table>
150<p>Note that in practice, the tenth byte could only encode bits 63 and 64
151since the maximum quantity to use this encoding is a 64-bit integer.</p>
152<p>The table below defines the encoding rules for type names used in the
153descriptions of blocks and fields in the next section. Any type name with
154the suffix <em>_vbr</em> indicate a quantity that is encoded using
155variable bit rate encoding as described above.</p>
156<table class="doc_table" >
157 <tr>
158 <th><b>Type</b></th>
159 <th align="left"><b>Rule</b></th>
160 </tr>
161 <tr>
162 <td>unsigned</td>
163 <td align="left">A 32-bit unsigned integer that always occupies four
164 consecutive bytes. The unsigned integer is encoded using LSB first
165 ordering. That is bits 2<sup>0</sup> through 2<sup>7</sup> are in the
166 byte with the lowest file offset (little endian).</td>
167 </tr><tr>
168 <td>uint_vbr</td>
169 <td align="left">A 32-bit unsigned integer that occupies from one to five
170 bytes using variable bit rate encoding.</td>
171 </tr><tr>
172 <td>uint64_vbr</td>
173 <td align="left">A 64-bit unsigned integer that occupies from one to ten
174 bytes using variable bit rate encoding.</td>
175 </tr><tr>
176 <td>int64_vbr</td>
177 <td align="left">A 64-bit signed integer that occupies from one to ten
178 bytes using variable bit rate encoding.</td>
179 </tr><tr>
180 <td>char</td>
181 <td align="left">A single unsigned character encoded into one byte</td>
182 </tr><tr>
183 <td>bit</td>
184 <td align="left">A single bit within a byte.</td>
185 </tr><tr>
186 <td>string</td>
187 <td align="left">A uint_vbr indicating the length of the character string
188 immediately followed by the characters of the string. There is no
189 terminating null byte in the string. Characters are interpreted as unsigned
190 char and are generally US-ASCII encoded.</td>
191 </tr><tr>
192 <td>data</td>
193 <td align="left">An arbitrarily long segment of data to which no
194 interpretation is implied. This is used for float, double, and constant
195 initializers.</td>
196 </tr>
197</table>
198</div>
199<!-- _______________________________________________________________________ -->
Reid Spencer50026612004-05-22 02:28:36 +0000[diff] [blame]200<div class="doc_subsection"><a name="align">Alignment</a> </div>
201<div class="doc_text">
202<p>To support cross-platform differences, the bytecode file is aligned on
203certain boundaries. This means that a small amount of padding (at most 3 bytes)
204will be added to ensure that the next entry is aligned to a 32-bit boundary.
205</p>
206</div>
207<!-- *********************************************************************** -->
208<div class="doc_section"> <a name="details">Detailed Layout</a> </div>
209<!-- *********************************************************************** -->
210<div class="doc_text">
Reid Spencerb39021b2004-05-23 17:05:09 +0000[diff] [blame^]211<p>This section provides the detailed layout of the LLVM bytecode file format.
212bit and byte level specifics.</p>
Reid Spencer50026612004-05-22 02:28:36 +0000[diff] [blame]213</div>
214<!-- _______________________________________________________________________ -->
215<div class="doc_subsection"><a name="notation">Notation</a></div>
216<div class="doc_text">
217 <p>The descriptions of the bytecode format that follow describe the bit
218 fields in detail. These descriptions are provided in tabular form. Each table
219 has four columns that specify:</p>
220 <ol>
221 <li><b>Byte(s)</b>. The offset in bytes of the field from the start of
222 its container (block, list, other field).<li>
223 <li><b>Bit(s)</b>. The offset in bits of the field from the start of
224 the byte field. Bits are always little endian. That is, bit addresses with
225 smaller values have smaller address (i.e. 2^0 is at bit 0, 2^1 at 1, etc.)
226 </li>
227 <li><b>Align?</b> Indicates if this field is aligned to 32 bits or not.
228 This indicates where the <em>next</em> field starts, always on a 32 bit
229 boundary.</li>
Reid Spencer939290f2004-05-22 05:56:41 +0000[diff] [blame]230 <li><b>Type</b>. The basic type of information contained in the field.</li>
Reid Spencer50026612004-05-22 02:28:36 +0000[diff] [blame]231 <li><b>Description</b>. Descripts the contents of the field.</li>
232 </ol>
233</div>
234<!-- _______________________________________________________________________ -->
235<div class="doc_subsection"><a name="blocktypes">Block Types</a></div>
236<div class="doc_text">
237 <p>The bytecode format encodes the intermediate representation into groups
238 of bytes known as blocks. The blocks are written sequentially to the file in
239 the following order:</p>
240<ol>
Reid Spencerb39021b2004-05-23 17:05:09 +0000[diff] [blame^]241 <li><a href="#signature">Signature</a>. This block contains the file signature
242 (magic number) that identifies the file as LLVM bytecode.</li>
243 <li><a href="#module">Module Block</a>. This is the top level block in a
244 bytecode file. It contains all the other blocks.</li>
Reid Spencer50026612004-05-22 02:28:36 +0000[diff] [blame]245 <li><a href="#gtypepool">Global Type Pool</a>. This block contains all the
246 global (module) level types.</li>
247 <li><a href="#modinfo">Module Info</a>. This block contains the types of the
248 global variables and functions in the module as well as the constant
249 initializers for the global variables</li>
250 <li><a href="#constants">Constants</a>. This block contains all the global
251 constants except function arguments, global values and constant strings.</li>
252 <li><a href="#functions">Functions</a>. One function block is written for
253 each function in the module. </li>
254 <li><a href="$symtab">Symbol Table</a>. The module level symbol table that
255 provides names for the various other entries in the file is the final block
256 written.</li>
257</ol>
258</div>
259<!-- _______________________________________________________________________ -->
Reid Spencerb39021b2004-05-23 17:05:09 +0000[diff] [blame^]260<div class="doc_subsection"><a name="signature">Signature Block</a> </div>
Reid Spencer50026612004-05-22 02:28:36 +0000[diff] [blame]261<div class="doc_text">
Reid Spencerb39021b2004-05-23 17:05:09 +0000[diff] [blame^]262<p>The signature block occurs in every LLVM bytecode file and is always first.
263It simply provides a few bytes of data to identify the file as being an LLVM
264bytecode file. This block is always four bytes in length and differs from the
265other blocks because there is no identifier and no block length at the start
266of the block. Essentially, this block is just the "magic number" for the file.
267<table class="doc_table_nw" >
Reid Spencer50026612004-05-22 02:28:36 +0000[diff] [blame]268 <tr>
269 <th><b>Byte(s)</b></th>
270 <th><b>Bit(s)</b></th>
271 <th><b>Align?</b></th>
Reid Spencer939290f2004-05-22 05:56:41 +0000[diff] [blame]272 <th><b>Type</b></th>
Reid Spencer50026612004-05-22 02:28:36 +0000[diff] [blame]273 <th align="left"><b>Field Description</b></th>
Reid Spencerb39021b2004-05-23 17:05:09 +0000[diff] [blame^]274 </tr><tr>
275 <td>00</td><td>-</td><td>No</td><td>char</td>
Reid Spencer939290f2004-05-22 05:56:41 +0000[diff] [blame]276 <td align="left">Constant "l" (0x6C)</td>
Reid Spencerb39021b2004-05-23 17:05:09 +0000[diff] [blame^]277 </tr><tr>
278 <td>01</td><td>-</td><td>No</td><td>char</td>
Reid Spencer939290f2004-05-22 05:56:41 +0000[diff] [blame]279 <td align="left">Constant "l" (0x6C)</td>
Reid Spencerb39021b2004-05-23 17:05:09 +0000[diff] [blame^]280 </tr><tr>
281 <td>02</td><td>-</td><td>No</td><td>char</td>
Reid Spencer939290f2004-05-22 05:56:41 +0000[diff] [blame]282 <td align="left">Constant "v" (0x76)</td>
Reid Spencerb39021b2004-05-23 17:05:09 +0000[diff] [blame^]283 </tr><tr>
284 <td>03</td><td>-</td><td>No</td><td>char</td>
Reid Spencer939290f2004-05-22 05:56:41 +0000[diff] [blame]285 <td align="left">Constant "m" (0x6D)</td>
Reid Spencer50026612004-05-22 02:28:36 +0000[diff] [blame]286 </tr>
Reid Spencerb39021b2004-05-23 17:05:09 +0000[diff] [blame^]287</table>
288</div>
289<!-- _______________________________________________________________________ -->
290<div class="doc_subsection"><a name="module">Module Block</a> </div>
291<div class="doc_text">
292<p>The module block contains a small pre-amble and all the other blocks in
293the file. Of particular note, the bytecode format number is simply a 28-bit
294monotonically increase integer that identifiers the version of the bytecode
295format. While the bytecode format version is not related to the LLVM release
296(it doesn't automatically get increased with each new LLVM release), there is
297a definite correspondence between the bytecode format version and the LLVM
298release.</p>
299<p>The table below shows the format of the module block header. The blocks it
300contains are detailed in other sections.</p>
301<table class="doc_table_nw" >
Reid Spencer50026612004-05-22 02:28:36 +0000[diff] [blame]302 <tr>
Reid Spencerb39021b2004-05-23 17:05:09 +0000[diff] [blame^]303 <th><b>Byte(s)</b></th>
304 <th><b>Bit(s)</b></th>
305 <th><b>Align?</b></th>
306 <th><b>Type</b></th>
307 <th align="left"><b>Field Description</b></th>
308 </tr><tr>
309 <td>04-07</td><td>-</td><td>No</td><td>unsigned</td>
310 <td align="left">Module Identifier (0x01)</td>
311 </tr><tr>
312 <td>08-11</td><td>-</td><td>No</td><td>unsigned</td>
313 <td align="left">Size of the module block in bytes</td>
314 </tr><tr>
315 <td>12-15</td><td>00</td><td>Yes</td><td>uint32_vbr</td>
316 <td align="left">Format Information</td>
317 </tr><tr>
318 <td>''</td><td>0</td><td>-</td><td>bit</td>
319 <td align="left">Big Endian?</td>
320 </tr><tr>
321 <td>''</td><td>1</td><td>-</td><td>bit</td>
322 <td align="left">Pointers Are 64-bit?</td>
323 </tr><tr>
324 <td>''</td><td>2</td><td>-</td><td>bit</td>
325 <td align="left">Has No Endianess?</td>
326 </tr><tr>
327 <td>''</td><td>3</td><td>-</td><td>bit</td>
328 <td align="left">Has No Pointer Size?</td>
329 </tr><tr>
330 <td>''</td><td>4-31</td><td>-</td><td>bit</td>
331 <td align="left">Bytecode Format Version</td>
332 </tr><tr>
333 <td>16-end</td><td>-</td><td>-</td><td>blocks</td>
334 <td align="left">The remaining bytes in the block consist
335 solely of other block types in sequence.</td>
Reid Spencer50026612004-05-22 02:28:36 +0000[diff] [blame]336 </tr>
337</table>
338</div>
339<!-- _______________________________________________________________________ -->
340<div class="doc_subsection"><a name="gtypepool">Global Type Pool</a> </div>
341<div class="doc_text">
Reid Spencerb39021b2004-05-23 17:05:09 +0000[diff] [blame^]342<p>The global type pool consists of type definitions. Their order of appearnce
343in the file determines their slot number (0 based). Slot numbers are used to
344replace pointers in the intermediate representation. Each slot number uniquely
345identifies one entry in a type plane (a collection of values of the same type).
346Since all values have types and are associated with the order in which the type
347pool is written, the global type pool <em>must</em> be written as the first
348block of a module. If it is not, attempts to read the file will fail because
349both forward and backward type resolution will not be possible.</p>
350<p>The type pool is simply a list of types definitions, as shown in the table
351below.</p>
352<table class="doc_table_nw" >
353 <tr>
354 <th><b>Byte(s)</b></th>
355 <th><b>Bit(s)</b></th>
356 <th><b>Align?</b></th>
357 <th><b>Type</b></th>
358 <th align="left"><b>Field Description</b></th>
359 </tr><tr>
360 <td>00-03</td><td>-</td><td>No</td><td>unsigned</td>
361 <td align="left">Type Pool Identifier (0x13)</td>
362 </tr><tr>
363 <td>04-07</td><td>-</td><td>No</td><td>unsigned</td>
364 <td align="left">Size in bytes of the symbol table block.</td>
365 </tr><tr>
366 <td>08-11<sup>1</sup></td><td>-</td><td>No</td><td>uint32_vbr</td>
367 <td align="left">Number of entries in type plane</td>
368 </tr><tr>
369 <td>12-15<sup>1</sup></td><td>-</td><td>No</td><td>uint32_vbr</td>
370 <td align="left">Type plane index for following entries</td>
371 </tr><tr>
372 <td>16-end<sup>1,2</sup></td><td>-</td><td>No</td><td>type</td>
373 <td align="left">Each of the type definitions.</td>
374 </tr><tr>
375 <td align="left" colspan="5"><sup>1</sup>Maximum length shown,
376 may be smaller<br><sup>2</sup>Repeated field.
377 </tr>
378</table>
Reid Spencer50026612004-05-22 02:28:36 +0000[diff] [blame]379</div>
380<!-- _______________________________________________________________________ -->
381<div class="doc_subsection"><a name="modinfo">Module Info</a> </div>
382<div class="doc_text">
Reid Spencerb39021b2004-05-23 17:05:09 +0000[diff] [blame^]383 <p>To be determined.</p>
Reid Spencer50026612004-05-22 02:28:36 +0000[diff] [blame]384</div>
385<!-- _______________________________________________________________________ -->
386<div class="doc_subsection"><a name="constants">Constants</a> </div>
387<div class="doc_text">
Reid Spencerb39021b2004-05-23 17:05:09 +0000[diff] [blame^]388 <p>To be determined.</p>
Reid Spencer50026612004-05-22 02:28:36 +0000[diff] [blame]389</div>
390<!-- _______________________________________________________________________ -->
391<div class="doc_subsection"><a name="functions">Functions</a> </div>
392<div class="doc_text">
Reid Spencerb39021b2004-05-23 17:05:09 +0000[diff] [blame^]393 <p>To be determined.</p>
Reid Spencer50026612004-05-22 02:28:36 +0000[diff] [blame]394</div>
395<!-- _______________________________________________________________________ -->
Reid Spencerb39021b2004-05-23 17:05:09 +0000[diff] [blame^]396<div class="doc_subsection"><a name="symtab">Symbol Table</a> </div>
Reid Spencer50026612004-05-22 02:28:36 +0000[diff] [blame]397<div class="doc_text">
Reid Spencerb39021b2004-05-23 17:05:09 +0000[diff] [blame^]398<p>A symbol table can be put out in conjunction with a module or a function.
399A symbol table is a list of type planes. Each type plane starts with the number
400of entries in the plane and the type plane's slot number (so the type can be
401looked up in the global type pool). For each entry in a type plane, the slot
402number of the value and the name associated with that value are written. The
403format is given in the table below. </p>
404<table class="doc_table_nw" >
405 <tr>
406 <th><b>Byte(s)</b></th>
407 <th><b>Bit(s)</b></th>
408 <th><b>Align?</b></th>
409 <th><b>Type</b></th>
410 <th align="left"><b>Field Description</b></th>
411 </tr><tr>
412 <td>00-03</td><td>-</td><td>No</td><td>unsigned</td>
413 <td align="left">Symbol Table Identifier (0x13)</td>
414 </tr><tr>
415 <td>04-07</td><td>-</td><td>No</td><td>unsigned</td>
416 <td align="left">Size in bytes of the symbol table block.</td>
417 </tr><tr>
418 <td>08-11<sup>1</sup></td><td>-</td><td>No</td><td>uint32_vbr</td>
419 <td align="left">Number of entries in type plane</td>
420 </tr><tr>
421 <td>12-15<sup>1</sup></td><td>-</td><td>No</td><td>uint32_vbr</td>
422 <td align="left">Type plane index for following entries</td>
423 </tr><tr>
424 <td>16-19<sup>1,2</sup></td><td>-</td><td>No</td><td>uint32_vbr</td>
425 <td align="left">Slot number of a value.</td>
426 </tr><tr>
427 <td>variable<sup>1,2</sup></td><td>-</td><td>No</td><td>string</td>
428 <td align="left">Name of the value in the symbol table.</td>
429 </tr>
430 <tr>
431 <td align="left" colspan="5"><sup>1</sup>Maximum length shown,
432 may be smaller<br><sup>2</sup>Repeated field.
433 </tr>
434</table>
Reid Spencer50026612004-05-22 02:28:36 +0000[diff] [blame]435</div>
436
437<!-- *********************************************************************** -->
438<hr>
439<address>
440 <a href="http://jigsaw.w3.org/css-validator/check/referer"><img
441 src="http://jigsaw.w3.org/css-validator/images/vcss" alt="Valid CSS!"></a>
442 <a href="http://validator.w3.org/check/referer"><img
443 src="http://www.w3.org/Icons/valid-html401" alt="Valid HTML 4.01!" /></a>
444
445 <a href="mailto:rspencer@x10sys.com">Reid Spencer</a> and
446 <a href="mailto:sabre@nondot.org">Chris Lattner</a><br>
447 <a href="http://llvm.cs.uiuc.edu">The LLVM Compiler Infrastructure</a><br>
448 Last modified: $Date$
449</address>
450</body>
451</html>
452<!-- vim: sw=2
453-->