Multimedia processor using variable length instructions with opcode specification of source operand as result of prior instruction

• US 5,864,704 A
• Filed: 10/10/1995
• Issued: 01/26/1999
• Est. Priority Date: 10/10/1995
• Status: Expired due to Term

• Current Assignee: ATI Technologies ULC (Advanced Micro Devices, Inc.)
• Original Assignee: Chromatic Research, Inc. (Advanced Micro Devices, Inc.)
• Inventors: Hung, Andy C., Battle, James Thomas, Purcell, Stephen C.
• Primary Examiner(s): Ellis, Richard L.
• Application Number:
  US08/541,565
• Time in Patent Office: 1,204 Days
• Field of Search: 395/382, 395/386, 395/388, 395/391, 395/800.24
• US Class Current: 712/24
• CPC Class Codes: G06F 9/30149 of variable length instructions
• CPC Class Codes: G06F 9/30181 Instruction operation extension or modification
• CPC Class Codes: G06F 9/30185 according to one or more bits in the instruction, e.g. prefix, sub-opcode
• CPC Class Codes: G06F 9/3826 Bypassing or forwarding of data results, e.g. locally between pipeline stages or within a pipeline stage
• CPC Class Codes: G06F 9/3853 of compound instructions

Abstract

A media engine is disclosed herein which incorporates into a single chip structure the seven multimedia functions of video, 2D graphics, 3D graphics, audio, FAX/modem, telephony, and video-conferencing. In accordance with the present invention, a media engine includes a signal processor which shares a memory with the CPU of the host computer and also includes a plurality of control modules each dedicated to one of the seven multi-media functions. The signal processor retrieves from this shared memory instructions placed therein by the host CPU and in response thereto causes the execution of such instructions via one of the on-chip control modules. The signal processor utilizes an instruction register having a movable partition which allows larger than typical instructions to be paired with smaller than typical instructions. The signal processor reduces demand for memory read ports by placing data into the instruction register where it may be directly routed to the arithmetic logic units for execution and, where the destination of a first instruction matches the source of a second instruction, by defaulting the source specifier of the second instruction to the result register of the ALU employed in the execution of the first instruction.

Claims

1. A multimedia circuit, comprising:

a plurality of arithmetic logic units (ALUs), each ALU having associated therewith a register to which an output datum of said ALU is stored; and

an instruction register, said instruction register accommodating first and second instructions, each instruction specifying an op-code, one of said ALUs and an operand op-code of said first instruction specifies as said operand source of said first instruction the result register associated with the ALU specified in said second instruction.

2. The multimedia circuit of claim 1 wherein said first instruction further specifies a second operand source, said opcode of said first instruction further specifying that the output datum in the register associated with the ALU specified in said opcode of said first instruction is to be stored in one of said first and second operand sources of said first instruction.

3. The multimedia circuit of claim 1 wherein said opcodes of said first and second instructions specifies whether said first and second instructions are executed sequentially or concurrently.

4. The multimedia circuit of claim 1 wherein said opcode of said first instruction indicates a number of times said first instruction is executed.

5. The multimedia circuit of claim 1 wherein said instruction register is eight bytes wide.

6. The multimedia circuit of claim 5 wherein said first and second instructions have different lengths.

7. In a microprocessor, a method comprising the steps of:

providing a plurality of arithmetic logic units (ALUs) each capable of executing an instruction and providing a datum in a register associated with each ALU;

providing an instruction register designed for holding a plurality of said instructions, each instruction specifying an opcode, an operand source and one of said ALUs for executing said instruction;

assigning each instruction in said instruction register to a selected one of said ALUs for execution;

executing said instructions of said instruction register in their respective assigned ALUs and providing a result datum for each of said instructions and determining whether said opcode of an instruction specifies that said result datum of said instruction is to be stored in said operand source.

8. The method of claim 7, further comprising the step of

determining in each instruction, based upon said opcode, whether or not said operand source specifies an immediate value, a register of an ALU, or a memory address.

9. The method of claim 7 further comprising the step of:

determining from said opcode of a first instruction in said instruction register and said opcode of a second instruction in said instruction register, whether said first and second instructions are executed sequentially or concurrently by a processor within said engine.

10. The method of claim 7 further comprising the step of

executing, based upon said opcode of an instruction in said instruction register, said instruction a predetermined number of times.