IPC분류정보
국가/구분 |
United States(US) Patent
등록
|
국제특허분류(IPC7판) |
|
출원번호 |
US-0267640
(2005-11-04)
|
등록번호 |
US-8738891
(2014-05-27)
|
발명자
/ 주소 |
- Karandikar, Ashish
- Gadre, Shirish
- Salek, Amir H.
|
출원인 / 주소 |
|
인용정보 |
피인용 횟수 :
1 인용 특허 :
149 |
초록
▼
A method for implementing command acceleration. The method includes receiving a first set of instructions from a first processor, wherein the first set of instructions are formatted in accordance with a microarchitecture of the first processor. The first set of instructions are translated into a sec
A method for implementing command acceleration. The method includes receiving a first set of instructions from a first processor, wherein the first set of instructions are formatted in accordance with a microarchitecture of the first processor. The first set of instructions are translated into a second set of instructions, wherein the second set of instructions are formatted in accordance with a microarchitecture of a second processor. The second set instructions are then transmitted to the second processor for execution by the second processor.
대표청구항
▼
1. A method for implementing command acceleration, said method comprising: receiving a first set of video processing instructions from a first processor, the first set of video processing instructions formatted in accordance with a microarchitecture of the first processor, and wherein said first set
1. A method for implementing command acceleration, said method comprising: receiving a first set of video processing instructions from a first processor, the first set of video processing instructions formatted in accordance with a microarchitecture of the first processor, and wherein said first set of video processing instructions comprise scalar instructions;translating the first set of video processing instructions into a second set of instructions, the second set of video processing instructions formatted in accordance with a microarchitecture of a second processor; andtransmitting the second set of video processing instructions to the second processor for execution by the second processor, wherein the first set of instructions for the first processor comprise packed instructions and the second set of instructions for the second processor comprise unpacked instructions. 2. The method of claim 1, wherein the first processor is a scalar processor, and wherein the first set of instructions are in accordance with a scalar microarchitecture of the scalar processor. 3. The method of claim 1, wherein the second processor is a vector processor, and wherein the second set of instructions are in accordance with a vector microarchitecture of the vector processor. 4. The method of claim 1, wherein the first set of instructions comprise at least one macro instruction. 5. The method of claim 4, wherein the at least one macro instruction is expanded into a plurality of instructions during the translating to produce the second set of instructions. 6. The method of claim 1, wherein the first set of instructions are in accordance with a scalar processor, and wherein the second set of instructions are configured to feed a plurality of inputs of a SIMD engine of the second processor. 7. A programmable command accelerator for improving instruction transfer from a scalar processor to a vector processor, comprising: an input for receiving a first set of video processing instructions from a first processor, the first set of video processing instructions formatted in accordance with a microarchitecture of the first processor, and wherein said first set of video processing instructions comprise scalar instructions;a programmable lookup table data structure for implementing translations for the first set of video processing instructions;a controller for accessing the lookup table data structure and for translating the first set of video processing instructions into a second set of video processing instructions, the second set of video processing instructions formatted in accordance with a microarchitecture of a second processor; andan output for transmitting the second set video processing instructions to the second processor for execution by the second processor, wherein the first set of instructions for the first processor comprise packed instructions and the second set of instructions for the second processor comprise unpacked instructions. 8. The command accelerator of claim 7, wherein the first processor is a scalar processor, and wherein the first set of instructions are in accordance with a scalar microarchitecture of the scalar processor. 9. The command accelerator of claim 7, wherein the second processor is a vector processor, and wherein in-the second set of instructions are in accordance with a vector microarchitecture of the vector processor. 10. The command accelerator of claim 7, wherein the first set of instructions comprise at least one macro instruction. 11. The command accelerator of claim 10, wherein the at least one macro instruction is expanded into a plurality of instructions during the translating to produce the second set of instructions. 12. The command accelerator of claim 7, wherein the first set of instructions are in accordance with a scalar processor, and wherein the second set of instructions are configured to feed a plurality of inputs of a SIMD engine of the second processor. 13. A system for executing video processing operations, comprising: a CPU; anda video processor coupled to the CPU, comprising: a scalar execution unit configured to execute scalar video processing operations;a vector execution unit configured to execute vector video processing operations; anda programmable command accelerator for transferring instructions from the scalar execution unit to the vector execution unit, comprising: an input for receiving a first set of video processing instructions from the scalar execution unit, the first set of video processing instructions formatted in accordance with a scalar microarchitecture of the scalar execution unit, and wherein said first set of video processing instructions comprise scalar instructions;a programmable lookup table data structure for implementing translations for the first set of video processing instructions;a controller for accessing the lookup table data structure and for translating the first set of video processing instructions into a second set of video processing instructions formatted in accordance with a vector microarchitecture; andan output for transmitting the second set of video processing instructions to the vector execution unit;wherein the first set of instructions for the first processor comprise packed instructions and the second set of instructions for the second processor comprise unpacked instructions. 14. The system of claim 13, wherein the first set of instructions comprise at least one macro instruction. 15. The system of claim 14, wherein the at least one macro instruction is expanded into a plurality of instructions during the translating to produce the second set of instructions. 16. The system of claim 13, wherein the first set of instructions are in accordance with the scalar execution unit, and wherein the second set of instructions are configured to feed a plurality of inputs of a SIMD engine of the vector execution unit. 17. The system of claim 13, wherein the scalar execution unit is configured to execute flow control algorithms and allocate a plurality of work packages to the vector execution unit via the programmable command accelerator. 18. The system of claim 13, wherein the vector execution unit is configured to function as a slave co-processor to the scalar execution unit. 19. The system of claim 13, wherein said packed instructions comprise instructions with redundant and nonvariable constants removed.
※ AI-Helper는 부적절한 답변을 할 수 있습니다.