System for executing instructions having flag for indicating direct or indirect specification of a length of operand data
First Claim
1. A graphics processor for executing a computer instruction set comprising a plurality of instructions each of which has an instruction opcode and at least one operand, wherein a length of all the instructions is equal and of a like predetermined length, the opcode corresponds to a type of calculation to be performed on the operand(s), each operand is data to be processed in the calculation or specifies an address of the data, a result of the calculation represents processed image data, and each instruction includes a length field that includes data specifying a number of items of data to be processed or, if the number exceeds a size of the length field, a predetermined location of a previously-allocated storage area at which the number is stored, and a flag indicating whether the number is stored in the storage area or the length field, whereby said processor for each instruction processes a corresponding number of data items to thereby facilitate processing of variable-length streams of data.
2 Assignments
0 Petitions
Accused Products
Abstract
The present invention discloses an image processor (224) for executing a computer instruction set (280, 290) in the form of an opcode (281), at least one operand (283-285) which is, or indicates the location of data to be processed. The data to be processed consists of a variable length stream of data and each instruction includes a length field (297) containing data specifying the number of items of data to be processed or, if that number exceeds the size of the length field, a predetermined location of a previously allocated storage area at which that number is stored.
189 Citations
7 Claims
- 1. A graphics processor for executing a computer instruction set comprising a plurality of instructions each of which has an instruction opcode and at least one operand, wherein a length of all the instructions is equal and of a like predetermined length, the opcode corresponds to a type of calculation to be performed on the operand(s), each operand is data to be processed in the calculation or specifies an address of the data, a result of the calculation represents processed image data, and each instruction includes a length field that includes data specifying a number of items of data to be processed or, if the number exceeds a size of the length field, a predetermined location of a previously-allocated storage area at which the number is stored, and a flag indicating whether the number is stored in the storage area or the length field, whereby said processor for each instruction processes a corresponding number of data items to thereby facilitate processing of variable-length streams of data.
-
4. A graphics processor for executing a computer instruction set comprising a plurality of instructions each of which has an instruction opcode and at least one operand, wherein a length of all of the instructions is equal and of a like predetermined length, the opcode corresponds to a type of calculation to be performed on the operand(s), each operand is data to be processed in the calculation or specifies an address of the data, a result of the calculation represents processed image data, and each instruction includes a length field containing data specifying a number of items of data to be processed or, if the number exceeds a size of the length field, a predetermined location of a previously-allocated storage area at which the number is stored, whereby said processor for each instruction processes a corresponding number of data items to thereby facilitate processing of variable-length streams of data.
-
5. A method of executing a computer instruction, wherein the computer instruction includes an instruction opcode and at least one operand, the opcode corresponds to a type of calculation to be performed on the at least one operand, each operand is data to be processed in the calculation or specifies an address of the data, and a result of the calculation represents processed image data, said method comprising the steps of:
-
determining a number of items to be processed by the instruction, wherein the instruction includes a length field that includes data specifying the number of items of data to be processed or, if the number exceeds a size of the length field, a predetermined location of a previously-allocated storage area at which the number is stored, and a flag indicating whether the number is stored in the storage area or the length field; and
processing the number of data items by the instruction to thereby facilitate processing of variable-length streams of data.
-
-
6. An apparatus for executing a computer instruction, wherein the computer instruction includes an instruction opcode and at least one operand, the opcode corresponds to a type of calculation to be performed on the at least one operand, each operand is data to be processed in the calculation or specifies an address of the data, and a result of the calculation represents processed image data, said apparatus comprising:
-
a determination circuit adapted to determine a number of items to be processed by the instruction, wherein the instruction includes a length field that includes data specifying the number of items of data to be processed or, if the number exceeds a size of the length field, a predetermined location of a previously-allocated storage area at which the number is stored, and a flag indicating whether the number is stored in the storage area or the length field; and
a processor circuit adapted to process the number of data items by the instruction to thereby facilitate processing of variable-length streams of data.
-
-
7. A computer-readable medium storing a computer program for executing a computer instruction, wherein the computer instruction including an instruction opcode and at least one operand, the opcode corresponds to a type of calculation to be performed on the at least one operand, each operand is data to be processed in the calculation or specifies an address of the data, and a result of the calculation represents processed image data, the computer program comprising:
-
code for determining a number of items to be processed by the instruction, wherein the instruction includes a length field that includes data specifying the number of items of data to be processed or, if the number exceeds a size of the length field, a predetermined location of a previously-allocated storage area at which the number is stored, and a flag indicating whether the number is stored in the storage area or the length field; and
code for processing the number of data items by the instruction to thereby facilitate processing of variable-length streams of data.
-
Specification