公开(公告)号：US10204052B2

公开(公告)日：2019-02-12

申请号：US14635163

申请日：2015-03-02

Applicant: Huawei Technologies Co., Ltd.

Inventor： Chenghong He ,  Yongbo Cheng ,  Kejia Lan

IPC: G06F12/00 ,  G06F12/0831 ,  G06F12/0817 ,  G06F12/12

Abstract: A directory maintenance method and apparatus are provided. The method includes sending, by a main memory according to a correspondence between a cache line in a directory and a cache, listening information to each cache corresponding to a cache line at a preset frequency; receiving, by each cache corresponding to the cache line, the listening information, and sending a listening response according to the listening information; and receiving, by the main memory, the listening response, and updating the directory according to the listening response, where the listening response includes a state of the cache line in the cache sending the listening response. The directory maintenance method and apparatus that are disclosed in the present invention can lower an impact of listening caused due to replacement on normal processing of a processor, and reduce degradation of system performance.

公开(公告)号：US20150189039A1

公开(公告)日：2015-07-02

申请号：US14581577

申请日：2014-12-23

Applicant: Huawei Technologies Co., Ltd.

Inventor： Yongbo Cheng ,  Chenghong He ,  Kejia Lan

IPC: H04L29/08 ,  G06F12/08

CPC classification number: G06F12/0831 ,  G06F12/0813 ,  G06F12/0815 ,  G06F2212/2542 ,  H04L67/2842

Abstract: A memory data access method and apparatus, and a system are provided. In the embodiments of the present invention, when it is determined, according to a preset rule, that memory data located on a remote node needs to be frequently accessed, the memory data located on the remote node is replicated to a memory of a local node, and then the memory data located on the remote node is accessed from the memory of the local node. Because a delay of accessing a memory of a processor in a local node is much less than a delay of accessing a memory of a remote processor, when memory data located on a remote node needs to be frequently accessed, a delay of reading the memory data located on the remote node may be significantly reduced by using the solution, thereby improving system performance.

Abstract translation: 提供存储器数据存取方法和装置，以及系统。 在本发明的实施例中，当根据预设规则确定需要频繁地访问位于远程节点上的存储器数据时，位于远程节点上的存储器数据被复制到本地节点的存储器 ，然后从本地节点的存储器访问位于远程节点上的存储器数据。 因为访问本地节点中的处理器的存储器的延迟远远小于访问远程处理器的存储器的延迟，所以当位于远程节点上的存储器数据需要被频繁访问时，读取存储器数据的延迟 通过使用该解决方案可以显着地减少位于远程节点上，从而提高系统性能。

公开(公告)号：US09513838B2

公开(公告)日：2016-12-06

申请号：US14132858

申请日：2013-12-18

Applicant: HUAWEI TECHNOLOGIES CO., LTD.

Inventor： Yongbo Cheng ,  Chenghong He ,  Kejia Lan

IPC: G06F3/06 ,  G06F11/16 ,  G06F11/20

CPC classification number: G06F3/065 ,  G06F3/0604 ,  G06F3/0659 ,  G06F3/0683 ,  G06F11/1666 ,  G06F11/20

Abstract: A method and an apparatus for processing a system command during memory backup. The method includes: acquiring a write address corresponding to a write operation command; if data corresponding to the write address has been read from a raw memory area but is not written to a backup memory area, mapping the write operation command to the raw memory area, and writing data to the write address in the raw memory area according to the write operation command; and deducting a set value from the write address to obtain an initial address to subsequently read data from the raw memory area. According to the embodiments of the present invention, a problem of system command blocking is solved during a memory backup operation, so that a system command is processed in a timely manner.

Abstract translation: 一种在存储器备份期间处理系统命令的方法和装置。 该方法包括：获取与写入操作命令相对应的写入地址; 如果从原始存储器区域读取与写入地址相对应的数据，但是未写入到备份存储器区域，则将写入操作命令映射到原始存储器区域，并将数据写入原始存储区域中的写入地址，根据 写操作命令; 并从写入地址中扣除设定值以获得初始地址，以便随后从原始存储区读取数据。 根据本发明的实施例，在存储器备份操作期间解决了系统命令阻塞的问题，从而及时处理系统命令。

公开(公告)号：US20150254184A1

公开(公告)日：2015-09-10

申请号：US14635163

申请日：2015-03-02

Applicant: Huawei Technologies Co., Ltd.

Inventor： Chenghong He ,  Yongbo Cheng ,  Kejia Lan

IPC: G06F12/08 ,  G06F12/12

CPC classification number: G06F12/0833 ,  G06F12/0817 ,  G06F12/12 ,  G06F2212/621 ,  G06F2212/622

Abstract: A directory maintenance method and apparatus are provided. The method includes sending, by a main memory according to a correspondence between a cache line in a directory and a cache, listening information to each cache corresponding to a cache line at a preset frequency; receiving, by each cache corresponding to the cache line, the listening information, and sending a listening response according to the listening information; and receiving, by the main memory, the listening response, and updating the directory according to the listening response, where the listening response includes a state of the cache line in the cache sending the listening response. The directory maintenance method and apparatus that are disclosed in the present invention can lower an impact of listening caused due to replacement on normal processing of a processor, and reduce degradation of system performance.

Abstract translation: 提供了一种目录维护方法和装置。 该方法包括：根据目录中的高速缓存行和高速缓存之间的对应关系，通过主存储器将预设频率的与高速缓存行相对应的每个高速缓存的信息发送到主存储器; 通过对应于所述高速缓存线的每个缓存器接收所述收听信息，以及根据所述收听信息发送收听响应; 并且通过主存储器接收收听响应，并根据收听响应更新目录，其中监听响应包括发送收听响应的缓存中的高速缓存行的状态。 在本发明中公开的目录维护方法和装置可以降低由替换引起的收听对处理器的正常处理的影响，并且降低系统性能的降低。

公开(公告)号：US20230261795A1

公开(公告)日：2023-08-17

申请号：US18147344

申请日：2022-12-28

Applicant: Huawei Technologies Co., Ltd.

Inventor： Tao He ,  Donghao Yu ,  Kejia Lan ,  Ying Li

IPC: H04L1/1607

CPC classification number: H04L1/1628

Abstract: A method for retransmitting data includes a transmitter receiving a retry request, where the retry request includes a sequence number of a first to-be-retransmitted flit. The transmitter reads data in a retry buffer based on the sequence number of the first to-be-retransmitted flit and retransmits the data, where the retry buffer includes flits of at least two types of data packets, and retransmission manners of the at least two types of data packets are different.

公开(公告)号：US20170315916A1

公开(公告)日：2017-11-02

申请号：US15581599

申请日：2017-04-28

Applicant: Huawei Technologies Co., Ltd.

Inventor： Yongbo Cheng ,  Kejia Lan ,  Chenghong He

IPC: G06F12/0817 ,  G06F12/0811

CPC classification number: G06F12/0817 ,  G06F12/0811 ,  G06F12/0822 ,  G06F12/0831 ,  G06F2212/621

Abstract: The present invention provide a directory management method including: receiving, by a first NC, a first data access request sent by a first processor on a local node; if the first NC determines that a first directory does not include a directory entry corresponding to a first access address and the first directory does not include an idle directory entry, clearing, by the first NC, directory content of a directory entry from the first directory; writing, by the first NC, directory content corresponding to the first data access request to the cleared directory entry; and if the first NC determines that a first snoop request is received, sending, by the first NC, a first snoop message to the processor on the local node.

公开(公告)号：US09594684B2

公开(公告)日：2017-03-14

申请号：US14732172

申请日：2015-06-05

Applicant: Huawei Technologies Co., Ltd.

Inventor： Kejia Lan ,  Yongbo Cheng ,  Chenghong He

IPC: G06F12/08

CPC classification number: G06F12/0831 ,  G06F2212/1016

Abstract: A method for temporarily storing data and a storage device is provided. The method for temporarily storing data is applied to the storage device, and the storage device includes a source agent and a target agent. The method includes: sending, by the source agent, a data obtaining request to the target agent; receiving, by the source agent, target data that is corresponding to the data obtaining request and is returned by the target agent; determining, by the source agent, whether a snooping request that is for the target data and sent by the target agent is received after the data obtaining request is sent and before the target data is received, where the snooping request indicates that the target agent is simultaneously processing an obtaining request from another source agent for the target data; and if the snooping request is received, discarding, by the source agent, the target data.

Abstract translation: 提供一种临时存储数据的方法和存储装置。 临时存储数据的方法被应用于存储装置，并且存储装置包括源代理和目标代理。 该方法包括：由源代理向目标代理发送数据获取请求; 由所述源代理接收与所述数据获取请求对应并由所述目标代理返回的目标数据; 由源代理确定在发送数据获取请求之后并且在接收到目标数据之前是否接收到用于目标数据并由目标代理发送的窥探请求，其中，窥探请求指示目标代理是 同时处理来自另一个源代理的针对目标数据的获取请求; 并且如果接收到窥探请求，则由源代理丢弃目标数据。

公开(公告)号：US20230018349A1

公开(公告)日：2023-01-19

申请号：US17935940

申请日：2022-09-28

Applicant: HUAWEI TECHNOLOGIES CO., LTD.

Inventor： Tianchi Hu ,  Kejia Lan ,  Wenkai Ling

IPC: H04L41/12

Abstract: A switching device is provided obtains information about devices connected to a plurality of ports of the switching device and determines response ports of the switching device based on the information about the devices connected to the plurality of ports. Different response ports of the switching device are connected to different devices, and each response port is configured to receive a management packet or a response to the management packet that is sent by a device connected to each response port in a device enumeration process.

公开(公告)号：US08914700B2

公开(公告)日：2014-12-16

申请号：US14025553

申请日：2013-09-12

Applicant: Huawei Technologies Co., Ltd.

Inventor： Yongbo Cheng ,  Chenghong He ,  Kejia Lan

IPC: H03M13/00 ,  H03M13/29 ,  G06F11/10

CPC classification number: H03M13/2906 ,  G06F11/1048

Abstract: A method according to an embodiment of the present disclosure comprising: receiving a read instruction transmitted by a host device, the read instruction including a first address; reading first data together with a first CRC code and a first ECC which are associated with the first data from a memory based on the first address; and performing error detection on the first data based on the first CRC code, and performing error correction on the first data based on the first ECC if an error is detected. With the embodiments of the disclosure, the CRC code with better capability of error detection is adopted to perform error detection on the data. If any error is detected, error correction is performed using the ECC. Thus, it is possible to overcome the problem as to insufficient capability of error detection of the ECC in the prior art, thereby improving the system performance.

Abstract translation: 根据本公开的实施例的方法，包括：接收由主机设备发送的读指令，所述读指令包括第一地址; 基于第一地址从第一数据读取与第一数据相关联的第一CRC码和第一ECC的第一数据; 以及基于所述第一CRC码对所述第一数据执行错误检测，以及如果检测到错误，则基于所述第一ECC执行对所述第一数据的纠错。 利用本公开的实施例，采用具有更好的错误检测能力的CRC码来对数据执行错误检测。 如果检测到任何错误，则使用ECC执行纠错。 因此，可以克服现有技术中ECC的错误检测能力不足的问题，从而提高系统性能。

公开(公告)号：US20140164724A1

公开(公告)日：2014-06-12

申请号：US14132858

申请日：2013-12-18

Applicant: HUAWEI TECHNOLOGIES CO., LTD.

Inventor： Yongbo Cheng ,  Chenghong He ,  Kejia Lan

IPC: G06F3/06

CPC classification number: G06F3/065 ,  G06F3/0604 ,  G06F3/0659 ,  G06F3/0683 ,  G06F11/1666 ,  G06F11/20

Abstract: A method and an apparatus for processing a system command during memory backup. The method includes: acquiring a write address corresponding to a write operation command; if data corresponding to the write address has been read from a raw memory area but is not written to a backup memory area, mapping the write operation command to the raw memory area, and writing data to the write address in the raw memory area according to the write operation command; and deducting a set value from the write address to obtain an initial address to subsequently read data from the raw memory area. According to the embodiments of the present invention, a problem of system command blocking is solved during a memory backup operation, so that a system command is processed in a timely manner.

Abstract translation: 一种在存储器备份期间处理系统命令的方法和装置。 该方法包括：获取与写入操作命令相对应的写入地址; 如果从原始存储器区域读取与写入地址相对应的数据，但是未写入到备份存储器区域，则将写入操作命令映射到原始存储器区域，并将数据写入原始存储区域中的写入地址，根据 写操作命令; 并从写入地址中扣除设定值以获得初始地址，以便随后从原始存储区读取数据。 根据本发明的实施例，在存储器备份操作期间解决了系统命令阻塞的问题，从而及时处理系统命令。