diff options
Diffstat (limited to 'docs/opcodes/opcode-24-filled-new-array.html')
| -rw-r--r-- | docs/opcodes/opcode-24-filled-new-array.html | 144 |
1 files changed, 144 insertions, 0 deletions
diff --git a/docs/opcodes/opcode-24-filled-new-array.html b/docs/opcodes/opcode-24-filled-new-array.html new file mode 100644 index 000000000..cba0c5d61 --- /dev/null +++ b/docs/opcodes/opcode-24-filled-new-array.html @@ -0,0 +1,144 @@ +<!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</title> +<link rel=stylesheet href="opcode.css"> +</head> + +<body> + +<h1>filled-new-array</h1> + +<h2>Purpose</h2> + +<p> +Construct an array of the given type and size, filling it with the supplied +contents. The type must be an array type. The array's contents must be +single-word (that is, no arrays of long or double). The constructed instance is +stored as a "result" in the same way that the method invocation instructions +store their results, so the constructed instance must be moved to a register +with a subsequent move-result-object instruction (if it is to be used). +</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>24 35c</td> + <td>filled-new-array {vD, vE, vF, vG, vA}, type@CCCC</td> + <td><code>B:</code> array size and argument word count (4 bits)<br/> + <code>C:</code> type index (16 bits)<br/> + <code>D..G, A:</code> argument registers (4 bits each)</td> +</tr> +</tbody> +</table> + +<h2>Constraints</h2> + +<ul> + <li> + The value B must not be greater than 5. + </li> + <li> + If B > 0, then D must be a valid register index in the current stack frame. + </li> + <li> + If B > 1, then E must be a valid register index in the current stack frame. + </li> + <li> + If B > 2, then F must be a valid register index in the current stack frame. + </li> + <li> + If B > 3, then G must be a valid register index in the current stack frame. + </li> + <li> + If B > 4, then A must be a valid register index in the current stack frame. + </li> + <li> + C must be a valid index into the type pool. + </li> + <li> + The type denoted by C must be a valid array type descriptor. + </li> + <li> + The element size of the type denoted by C must be no larger than 32 bits. + </li> + <li> + If the element type is a primitive type, then all actual arguments + (vD .. vA, depending on B) must be primitive, too. + </li> + <li> + If the element type is a reference type, then all actual arguments + (vD .. vA, depending on B) must be references, too. + </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 C. + </li> + <li> + If T is a reference type and it 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 array R of type T and length B. 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> + If B > 0 then R[0] = vD + </li> + <li> + If B > 1 then R[1] = vE + </li> + <li> + If B > 2 then R[2] = vF + </li> + <li> + If B > 3 then R[3] = vG + </li> + <li> + If B > 4 then R[4] = vA + </li> + </ul> + </li> + <li> + No reference to R is stored in any register. Instead, R 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 vB < 0 + </li> + <li> + All exceptions that are possible during class loading can occur. + </li> + <li> + All exceptions that are possible during instantiation can occur. + </li> +</ul> + +</body> +</html> |