How to use information mechanism to realize process mutual exclusion, process synchronization and precursor relationship

Before that, you can refer to this article 《 operating system 》- A preliminary understanding of semaphore mechanism Have a general understanding of the information base

Semaphore base realizes mutual exclusion of processes
There are several steps to realize the mutual exclusion of processes ：
1、 Analyze the key activities of concurrent processes , Dividing critical zone （ Such as ： Access to critical resource printers should be placed in the critical area ）
2、 Set up Mutex semaphore mutex, The initial value is 0（ Semaphore mutex Express “ Places to enter the critical zone ”）
3、 In the entry area P(mutex)—— Application resources
4、 In the exit area V(mutex)—— Release resources


Be careful ： Yes Different critical resources need Set different amounts of mutually exclusive information
P、V Operations must appear in pairs . The lack of P Operation will result in critical resources not being mutually exclusive , The lack of V The operation will cause the resource to be accessed and cannot be released , The waiting process will never be awakened

---

The information mechanism realizes process synchronization
Process synchronization ： Let the concurrent processes advance orderly as required
For example, in the implementation process 1 Must be better than process 2 First access critical resources , This process is called process synchronization when it is pushed forward according to certain requirements , Let the asynchronous and concurrent processes cooperate with each other 、 To advance in an orderly way
Realizing process synchronization can be divided into the following steps ：
1、 Analyze where process synchronization is needed , must One before and one after Perform two operations （ Or two sentences of code ）
2、 Set up Synchronous semaphore S, The initial value is 0
3、 stay “ Front operation ” After performing V(S)
4、 stay “ Post operation ” Before execution P(S)


understand
Semaphore S Represents a resource , At first, there was no such resource .P2 To use this resource will be locked in P(S) Operate here , until P1 After execution V(S) take S The number of add 1 after P2 To be carried out .
If you execute to V(S) operation , be S++ after S=1. Then when the execution reaches P(S) In operation , because S=1, Indicates that there are available resources , Will execute S–,S The value of changes back to 0,P2 The process will not execute block The original language , Instead, continue to execute the code 4.
If you execute to P(S) operation , because S=0,S-- after S=-1, Indicates that no resources are available at this time , therefore P During the operation block The original language , Active request blocking . Then when the code is executed 2, Then execute V(S) operation , S++, send S Change back 0, Because there are processes in the blocking queue corresponding to the semaphore , So it will V In operation wakeup The original language , Wake up the P2 process . such P2 You can continue to execute Code 4 了

---

The information mechanism realizes the precursor relationship
What is the implementation of precursor relationships, for example
example ： process P1 There is a code in S1,P2 There is a code in S2 ,P3 There is a code in S3 …… P6 There is a code in S6. These codes are required to be executed in the order shown in the following precursor diagram ：

In fact, each pair of precursor relationships is a process synchronization relationship ( We need to ensure that one before another ), Therefore, the implementation of precursor relationship can be divided into the following steps ：
1、 want Set a synchronization amount for each pair of precursor relationships
2、 stay “ Front operation ” Then execute the corresponding synchronization information V operation
3、 stay “ Post operation ” Previously, execute the corresponding synchronization information P operation

Content review ：