Providing memory training of dynamic random access memory (DRAM) systems using port-to-port loopbacks, and related methods, systems, and apparatuses
원문보기
IPC분류정보
국가/구분
United States(US) Patent
등록
국제특허분류(IPC7판)
G11C-007/10
G11C-029/02
출원번호
US-0589145
(2015-01-05)
등록번호
US-9767868
(2017-09-19)
발명자
/ 주소
Srinivas, Vaishnav
Brunolli, Michael Joseph
Chun, Dexter Tamio
West, David Ian
출원인 / 주소
QUALCOMM Incorporated
대리인 / 주소
Withrow & Terranova, PLLC
인용정보
피인용 횟수 :
0인용 특허 :
3
초록▼
Providing memory training of dynamic random access memory (DRAM) systems using port-to-port loopbacks, and related methods, systems, and apparatuses are disclosed. In one aspect, a first port within a DRAM system is coupled to a second port via a loopback connection. A training signal is sent to the
Providing memory training of dynamic random access memory (DRAM) systems using port-to-port loopbacks, and related methods, systems, and apparatuses are disclosed. In one aspect, a first port within a DRAM system is coupled to a second port via a loopback connection. A training signal is sent to the first port from a System-on-Chip (SoC), and passed to the second port through the loopback connection. The training signal is then returned to the SoC, where it may be examined by a closed-loop training engine of the SoC. A training result corresponding to a hardware parameter may be recorded, and the process may be repeated until an optimal result for the hardware parameter is achieved at the closed-loop training engine. By using a port-to-port loopback configuration, the DRAM system parameters regarding timing, power, and other parameters associated with the DRAM system may be trained more quickly and with lower boot memory usage.
대표청구항▼
1. A method for providing memory training for a dynamic random access memory (DRAM) system, comprising: disabling memory operations on a first port of the DRAM system and a second port of the DRAM system of a plurality of ports of the DRAM system;configuring the first port of the DRAM system and the
1. A method for providing memory training for a dynamic random access memory (DRAM) system, comprising: disabling memory operations on a first port of the DRAM system and a second port of the DRAM system of a plurality of ports of the DRAM system;configuring the first port of the DRAM system and the second port of the DRAM system to communicate via a loopback connection;receiving, by the first port of the DRAM system, a training signal from a System-on-Chip (SoC);providing, by the first port of the DRAM system, the training signal to the second port of the DRAM system via the loopback connection; andproviding, by the second port of the DRAM system, the training signal to the SoC. 2. The method of claim 1, wherein: receiving the training signal comprises receiving the training signal from a closed-loop training engine of the SoC; andproviding the training signal to the SoC comprises providing the training signal to the closed-loop training engine of the SoC. 3. The method of claim 2, wherein the training signal is one of one or more incremental training signals provided by the closed-loop training engine; the method further comprising: determining one or more training results corresponding to the one or more incremental training signals; anddetermining a hardware parameter for the DRAM system based on the one or more training results. 4. The method of claim 3, wherein determining the hardware parameter comprises determining a median between a first-failing incremental training signal of the one or more incremental training signals and a last-failing incremental training signal of the one or more incremental training signals. 5. The method of claim 3, wherein determining the one or more training results corresponding to the one or more incremental training signals comprises conducting a parameter sweep. 6. The method of claim 1, wherein the training signal comprises one or more of a timing training parameter and a voltage training parameter. 7. The method of claim 1, wherein the first port and the second port are adjacent to each other within the DRAM system. 8. The method of claim 1, wherein the loopback connection is bi-directional. 9. The method of claim 1, wherein each of the first port and the second port comprises a multiplexer (MUX) operative to select the loopback connection. 10. A system for providing memory training for a dynamic random access memory (DRAM) system, comprising: a System-on-Chip (SoC) communicatively coupled to the DRAM system; andthe DRAM system comprising a plurality of ports, comprising a first port and a second port communicatively coupled via a loopback connection;wherein: the SoC is configured to: disable memory operations on the first port of the DRAM system and the second port of the DRAM system; andconfigure the first port of the DRAM system and the second port of the DRAM system to communicate via the loopback connection; andthe DRAM system is configured to: receive, by the first port of the DRAM system, a training signal from the SoC;provide, by the first port of the DRAM system, the training signal to the second port of the DRAM system via the loopback connection; andprovide, by the second port of the DRAM system, the training signal to the SoC. 11. The system of claim 10, wherein the SoC comprises a closed-loop training engine; wherein the DRAM system is further configured to: receive the training signal from the closed-loop training engine of the SoC; andprovide the training signal to the closed-loop training engine of the SoC. 12. The system of claim 11, wherein the closed-loop training engine is configured to: provide one or more incremental training signals comprising the training signal;determine one or more training results corresponding to the one or more incremental training signals; anddetermine a hardware parameter for the DRAM system based on the one or more training results. 13. The system of claim 12, wherein the closed-loop training engine is configured to determine the hardware parameter by determining a median between a first-failing incremental training signal of the one or more incremental training signals and a last-failing incremental training signal of the one or more incremental training signals. 14. The system of claim 12, wherein the closed-loop training engine is configured to determine the one or more training results corresponding to the one or more incremental training signals by conducting a parameter sweep. 15. The system of claim 10, wherein the DRAM system is configured to receive the training signal comprising one or more of a timing training parameter and a voltage training parameter. 16. The system of claim 10, wherein the first port and the second port are adjacent within the DRAM system. 17. The system of claim 10, wherein the loopback connection is bi-directional. 18. The system of claim 10, wherein each of the first port and the second port comprises a multiplexer (MUX) configured to select the loopback connection. 19. An apparatus comprising a dynamic random access memory (DRAM) system, configured to: disable memory operations on a first port of the DRAM system and a second port of the DRAM system of a plurality of ports of the DRAM system;configure the first port of the DRAM system and the second port of the DRAM system to communicate via a loopback connection;receive, by the first port of the DRAM system, a training signal from a System-on-Chip (SoC);provide, by the first port of the DRAM system, the training signal to the second port of the DRAM system via the loopback connection; andprovide, by the second port of the DRAM system, the training signal to the SoC.
연구과제 타임라인
LOADING...
LOADING...
LOADING...
LOADING...
LOADING...
이 특허에 인용된 특허 (3)
Cho, Dongsik, Memory system and SoC including linear addresss remapping logic.
※ AI-Helper는 부적절한 답변을 할 수 있습니다.