filled-new-array/range

Purpose

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.

Details

Op & Format	Mnemonic / Syntax	Arguments
25 3rc	filled-new-array/range {vCCCC .. vNNNN}, type@BBBB	`A:` array size and argument word count (8 bits) `B:` type index (16 bits) `C:` first argument register (16 bits) `N = A + C - 1`

Constraints

vN must be a valid register index in the current stack frame (this means everything below vN is valid, too).
For all values I in the interval [C .. N] the following must hold:
- vI must not be part of a register pair
- If the array type is a simple type, vI must be a simple type, too.
- If the array type is a reference type, vI must be a reference type, too.
B must be a valid index into the type pool.
The type denoted by B must be an array type.
The element size of the type denoted by B must be no larger than 32 bits.

Behavior

An attempt is made to get a reference to the type T the name of which is contained in type pool entry B.
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.
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.
The elements of R are filled according to the following rules:
- J[0] = vC
- J[1] = v(C+1)
- ...
- J[vA] = vN
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.

Exceptions

NegativeArraySizeException if vA < 0