System and method providing efficient packing of unaligned data

A method comprising:

• storing, by processing circuitry of a computing device, multiple data elements of a non-power-of-two bit width in a plurality of registers, wherein a register has a power-of-two bit width, by:
• storing a first bit of a multiple data element in a first register;
• storing a second bit of the multiple data element in a second register;
• and
• storing subsequent bits of the multiple data elements in subsequent registers.

At least one computer-readable medium having stored thereon instructions which, when executed, cause a computing device to perform operations comprising:

• storing, by processing circuitry of the computing device, multiple data elements of a non-power-of-two bit width in a plurality of registers, wherein a register has a power-of-two bit width, by:
• storing a first bit of a multiple data element in a first register;
• storing a second bit of the multiple data element in a second register;
• and
• storing subsequent bits of the multiple data elements in subsequent registers.