blob: 944a8b164fa71f1ed643adfdc010e4e5c237dd0c [file] [log] [blame]
<!DOCTYPE html PUBLIC "-//W3C//DTD XHTML 1.0 Strict//EN" "http://www.w3.org/TR/xhtml1/DTD/xhtml1-strict.dtd">
<html>
<head>
<title>filled-new-array/range</title>
<link rel=stylesheet href="opcode.css">
</head>
<body>
<h1>filled-new-array/range</h1>
<h2>Purpose</h2>
<p>
Construct an array of the given type and size, filling it with the supplied
contents. Clarifications and restrictions are the same as filled-new-array,
described above.
</p>
<h2>Details</h2>
<table class="instruc">
<thead>
<tr>
<th>Op &amp; Format</th>
<th>Mnemonic / Syntax</th>
<th>Arguments</th>
</tr>
</thead>
<tbody>
<tr>
<td>25 3rc</td>
<td>filled-new-array/range {vCCCC .. vNNNN}, type@BBBB</td>
<td><code>A:</code> array size and argument word count (8 bits)<br/>
<code>B:</code> type index (16 bits)<br/>
<code>C:</code> first argument register (16 bits)<br/>
<code>N = A + C - 1</code></td>
</tr>
</tbody>
</table>
<h2>Constraints</h2>
<ul>
<li>
vN must be a valid register index in the current stack frame (this means
everything below vN is valid, too).
</li>
<li>
For all values I in the interval [C .. N] the following must hold:
<ul>
<li>
vI must not be part of a register pair
</li>
<li>
If the array type is a simple type, vI must be a simple type, too.
</li>
<li>
If the array type is a reference type, vI must be a reference type, too.
</li>
</ul>
</li>
<li>
B must be a valid index into the type pool.
</li>
<li>
The type denoted by B must be an array type.
</li>
<li>
The element size of the type denoted by B must be no larger than 32 bits.
</li>
</ul>
<h2>Behavior</h2>
<ul>
<li>
An attempt is made to get a reference to the type T the name of which is
contained in type pool entry B.
</li>
<li>
If T has not been loaded and resolved before, it is being loaded and
resolved. All exceptions that are possible during class loading can occur at
this point.
</li>
<li>
An attempt is made to create a new instance J of type T and length vA. All
exceptions that are possible during instantiation can occur at this point.
</li>
<li>
The elements of R are filled according to the following rules:
<ul>
<li>
J[0] = vC
</li>
<li>
J[1] = v(C+1)
</li>
<li>
...
</li>
<li>
J[vA] = vN
</li>
</ul>
</li>
<li>
No reference to J is stored in any register. Instead, J can be accessed by a
move-result-object instruction immediately following this filled-new-array
instruction.
</li>
</ul>
<h2>Exceptions</h2>
<ul>
<li>
NegativeArraySizeException if vA < 0
</li>
</ul>
</body>
</html>