公开(公告)号：US09208100B2

公开(公告)日：2015-12-08

申请号：US13682891

申请日：2012-11-21

Applicant: Huawei Technologies Co., Ltd.

Inventor： Ce Yao ,  Yafei Zhao ,  Jianrong Xu

IPC: G06F12/08 ,  G06F12/12

CPC classification number: G06F12/0891 ,  G06F12/082 ,  G06F12/0831 ,  G06F12/12

Abstract: The present invention provides a directory replacement method and device. An HA receives a data access request including a first address from a first CA, if a designated storage where a directory is located is entirely occupied by the directory, and a first directory entry corresponding to the first address is not in the directory, the HA selects a second directory entry from the directory, deletes it and adds the first directory entry into the directory; before the HA replaces the directory entry in the directory, no matter what status (for example, I status, S status or A status) a share status of a cache line corresponding to an address in the directory entry to be replaced is, the HA does not need to request a corresponding CA to perform an invalidating operation on data, but directly replaces the directory entry in the directory, thereby improving replacement efficiency.

Abstract translation: 本发明提供一种目录替换方法和装置。 如果目录所在的指定存储器被目录完全占用，则HA从第一CA接收包括第一地址的数据访问请求，并且与该第一地址对应的第一目录条目不在该目录中，HA 从目录中选择第二个目录条目，将其删除并将第一个目录条目添加到目录中; 在HA替换目录中的目录条目之前，无论什么状态（例如，状态，S状态或A状态）与要替换的目录条目中的地址对应的高速缓存行的共享状态是HA 不需要请求相应的CA对数据​​执行无效操作，而是直接替换目录中的目录条目，从而提高替换效率。

公开(公告)号：US10324646B2

公开(公告)日：2019-06-18

申请号：US15066623

申请日：2016-03-10

Applicant: Huawei Technologies Co., Ltd.

Inventor： Gongyi Wang ,  Ben Chen ,  Yafei Zhao

IPC: G06F3/06 ,  G06F9/54 ,  G06F12/0813 ,  G06F12/0817 ,  G06F12/0831

Abstract: A node controller-based request responding method and node controller, where the method includes receiving, by a first node controller, a first packet, acquiring an information directory, and querying, in the information directory, whether a memory address requested by the first packet is occupied by a second node controller, and when the memory address requested by the first packet is occupied by the second node controller, querying node presence information to determine whether the second node controller exists, and when it is determined that the second node controller does not exist, generating and sending an invalid response packet.

公开(公告)号：US20160196087A1

公开(公告)日：2016-07-07

申请号：US15066623

申请日：2016-03-10

Applicant: Huawei Technologies Co., Ltd.

Inventor： Gongyi Wang ,  Ben Chen ,  Yafei Zhao

IPC: G06F3/06

CPC classification number: G06F3/0638 ,  G06F3/0604 ,  G06F3/0673 ,  G06F9/54 ,  G06F12/0813 ,  G06F12/0817 ,  G06F12/0831 ,  G06F2212/1048

Abstract: A node controller-based request responding method and node controller, where the method includes receiving, by a first node controller, a first packet, acquiring an information directory, and querying, in the information directory, whether a memory address requested by the first packet is occupied by a second node controller, and when the memory address requested by the first packet is occupied by the second node controller, querying node presence information to determine whether the second node controller exists, and when it is determined that the second node controller does not exist, generating and sending an invalid response packet.

Abstract translation: 一种基于节点控制器的请求响应方法和节点控制器，其中所述方法包括由第一节点控制器接收第一分组，获取信息目录，以及在所述信息目录中查询由所述第一分组请求的存储器地址 被第二节点控制器占用，并且当由第一分组请求的存储器地址被第二节点控制器占用时，查询节点存在信息以确定第二节点控制器是否存在，以及何时确定第二节点控制器执行 不存在，生成和发送无效的响应数据包。

公开(公告)号：US20130111150A1

公开(公告)日：2013-05-02

申请号：US13708670

申请日：2012-12-07

Applicant: Huawei Technologies Co., Ltd.

Inventor： Yafei Zhao ,  Ruoxing Dai ,  Xiaowei Chu

IPC: G06F12/08

CPC classification number: G06F12/0891 ,  G06F12/0808 ,  G06F12/0828 ,  G06F12/0833

Abstract: A method for preventing deadlock of a node controller includes receiving, by an NC of a current node, a request message from any node, and writing the request message into a processing queue; performing invalidation processing on the cache data block that includes the system address and is cached on the other node, so that the NC directly returns, upon receiving a first interception message transferred by an HA, a feedback message to the HA, thereby avoiding that the first interception message is written into the processing queue and is blocked by the request message; and the feedback message is used for indicating that the system address cached on the other node is invalid; and transferring, by the NC, to the HA, the request message written into the processing queue.

Abstract translation: 一种防止节点控制器死锁的方法包括：从当前节点的NC接收来自任何节点的请求消息，并将请求消息写入处理队列; 对包括系统地址的高速缓存数据块执行无效处理，并且缓存在另一个节点上，使得NC在接收到由HA传送的第一个拦截消息时直接返回到HA的反馈消息，从而避免了 第一个拦截消息被写入处理队列并被请求消息阻塞; 反馈消息用于指示缓存在另一个节点上的系统地址无效; 并由NC向HA转发写入处理队列的请求消息。