AMBA CHI协议学习笔记4 - Ordering，Hazard

【本系列文章是我自己对CHI协议的学习笔记，仅供参考，欢迎指正。】

【文中的红色字体是我自己的标注，本文参考了 谷公子 作品，特此鸣谢】

谷公子

一、ordering，

1、Multi-copy atomicity 就是对同地址的写操作串行化，由放置在HN的POS完成排序。

The memory model used in this specification requires multi-copy atomicity. All compliant components must ensure that all write-type requests are multi-copy atomic. A write is defined as multi-copy atomic if both of the following conditions are true:

• All writes to the same location are serialized, that is, they are observed in the same order by all Requesters,
although some Requesters might not observe all of the writes.
• A read of a location does not return the value of a write until all Requesters observe that write.
In this specification, two addresses are considered to be the same with respect to coherence, observability, and hazarding if their cache line addresses and Physical Address Space (PAS) attributes are the same.

2、Completion Response and Ordering  就是对同地址或者同端点地址范围的操作 用 Comp，RespSepData，CompData响应来保障其被观测到。也就是必须等到了才能next。

 3、Completion acknowledgement

 CompletionAcknowledgment 负责 Requester发送的transactions和其它Requester transactions产生的snoop之间的保序关系。

 4、CHI协议也定义了RN和HN、HN-I和SN-I之间的requests保序的机制，HN-F和SN-F之间的order域是用于获得request accepted acknowledge

Order域只有在以下操作中才能置为非0值：
ReadNoSnp，ReadNoSnpSep，ReadOnce*，WriteNoSnp，WriteUnique，Atomic

 5、RespSepData and DataSepResp

就是数据必须发完，话必须说完的意思。

6、Stream Ordering WriteUnique transactions

    Optimized Streaming Ordered WriteUniques

就是写操作的流水化，不必等待comp。

二、Hazard

1、At the RN-F node

如果RN-F发送一个访问相同cacheline正在pending的request，并且该pending request还没有接收到任何Data response packet：snoop request必须要正常处理。

如果RN-F发送一个访问相同cacheline正在pending的request，并且该pending request已经接收到至少一个Data response packet，RN-F必须等待所有的Data response packets收齐后，才能处理snoop request。

 

 

2、At the ICN(HN-F) node

HN-F会对同地址的transactions进行保序，是通过对transaction responses和snoop transactions排序发送给Requesters来保证的。