Intel ITANIUM ARCHITECTURE - SOFTWARE DEVELOPERS MANUAL VOLUME 1 REV 2.3 Manual page 1760
Hide thumbs
Also See for ITANIUM ARCHITECTURE - SOFTWARE DEVELOPERS MANUAL VOLUME 1 REV 2.3:
Table of Contents
Advertisement
PUNPCKLBW/PUNPCKLWD/PUNPCKLDQ—Unpack Low Packed Data
Opcode
0F 60 /r
0F 61 /r
0F 62 /r
Description
Unpacks and interleaves the low-order data elements (bytes, words, or doublewords) of
the destination and source operands into the destination operand (see
The destination operand must be an MMX technology register; the source operand may
be either an MMX technology register or a memory location. When source data comes
from an MMX technology register, the upper 32 bits of the register are ignored. When
the source data comes from a memory, only 32-bits are accessed from memory.
Figure 3-23.
The PUNPCKLBW instruction interleaves the four low-order bytes of the source operand
and the four low-order bytes of the destination operand and writes them to the
destination operand.
The PUNPCKLWD instruction interleaves the two low-order words of the source operand
and the two low-order words of the destination operand and writes them to the
destination operand.
The PUNPCKLDQ instruction interleaves the low-order doubleword of the source
operand and the low-order doubleword of the destination operand and writes them to
the destination operand.
If the source operand is all zeros, the result (stored in the destination operand)
contains zero extensions of the high-order data elements from the original value in the
destination operand. With the PUNPCKLBW instruction the low-order bytes are zero
extended (that is, unpacked into unsigned words), and with the PUNPCKLWD
instruction, the low-order words are zero extended (unpacked into unsigned
doublewords).
4:458
Instruction
PUNPCKLBW mm,
mm/m32
PUNPCKLWD mm,
mm/m32
PUNPCKLDQ mm, mm/m32 Interleave low-order doublewords from mm and mm/m64
Low-order Unpacking and Interleaving of Bytes with the
PUNPCKLBW Instruction
PUNPCKLBW mm, mm/m32
mm/m32
2 2
2 2
3
2
1
2 1
3
3
Volume 4: IA-32 Intel
Description
Interleave low-order bytes from mm and mm/m64 into
mm .
Interleave low-order words from mm and mm/m64 into
mm .
into mm .
mm
1 1
1 1 1 1
0
7
6
5
4
3
2 1 2 1
2 1
2
2
1
1
0
0
mm
®
MMX™ Technology Instruction Reference
Figure
3-23).
1 1
2
1
0
3006032
Advertisement
Table of Contents
