diff options
Diffstat (limited to 'docs/opcodes/opcode-25-filled-new-array-range.html')
-rw-r--r-- | docs/opcodes/opcode-25-filled-new-array-range.html | 125 |
1 files changed, 125 insertions, 0 deletions
diff --git a/docs/opcodes/opcode-25-filled-new-array-range.html b/docs/opcodes/opcode-25-filled-new-array-range.html new file mode 100644 index 0000000..2ee7505 --- /dev/null +++ b/docs/opcodes/opcode-25-filled-new-array-range.html @@ -0,0 +1,125 @@ +<!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 & 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> |