Update Lec 15/16 on IFDS & Soundiness.

taylorgag · Dec 31, 2020 · ed39031 · ed39031
1 parent 63481a9
commit ed39031
Show file tree

Hide file tree

Showing 94 changed files with 475 additions and 46 deletions.
diff --git a/README.assets/image-20201231205814145.png b/README.assets/image-20201231205814145.png
diff --git a/README.md b/README.md
@@ -30,7 +30,13 @@ Getting started with static program analysis. Read this and start writing your f
 
 1. Oct, 2020. 设立Repo，一个月内解锁Star、Fork和PR。
 2. Nov. 将IR与Data Flow Analysis的相关内容暂时移出仓库，更新七至十课——Interprocedural Analysis、Pointer Analysis-Introduction and Foundations。
-3. Dec. 更新十一和十二两课——Context Sensitive Pointer Analysis，指针分析大结局。🥳 
+3. Dec. 更新十一和十二两课——Context Sensitive Pointer Analysis，指针分析大结局。🥳 更新十三十四课介绍指针分析的安全应用Taint Analysis和使用Datalog实现声明式指针分析算法。 更新十五十六课介绍IFDS分析框架和Soundiness。
+
+图文的主体部分更新完毕，撒花~！
+
+后续会考虑录制一些视频补充对动态例子的讲解。
+
+![感谢读者支持 :P](README.assets/image-20201231205814145.png)
 
 ## 这一《静态程序分析》教程对谁有用？
 

diff --git a/SUMMARY.md b/SUMMARY.md
@@ -4,7 +4,7 @@
 
 ## 前言 <a id="ch0"></a>
 
-* [写在前面](ch0/README.md)
+* [写在前面](ch0/ch0.md)
   * [为什么是这本书？](ch0/00-01-why-this-book.md)
   * [资料来源与版权信息](ch0/00-02-sources-and-license.md)
 
@@ -26,10 +26,13 @@
   * [上下文敏感分析（上）](ch3/context-sensitivity/03-04-context-sensitivity.md)
   * [上下文敏感分析（下）](ch3/context-sensitivity/03-05-cs2.md)
 
-## 指针分析应用 <a id="ch4"></a>
+## 指针分析应用与声明式实现 <a id="ch4"></a>
 
-* [安全应用——污点分析](ch4/04-01-security.md)
+* [应用——污点分析](ch4/04-01-security.md)
+* [实现——声明式指针分析](ch4/04-02-Datalog-Based-PA.md)
 
-## 指针分析的另一种样子 <a id="ch5"></a>
+## 其他话题 <a id="ch5"></a>
+
+* [另一种静态分析框架——IFDS](ch5/05-01-IFDS.md)
+* [从Soundness到Soundiness](ch5/05-02-Soundiness.md)
 
-* [声明式指针分析算法](ch5/05-01-Datalog-Based-PA.md)
diff --git a/ch0/README.md → ch0/ch0.md b/ch0/README.md → ch0/ch0.md
diff --git a/ch1/ch1.md b/ch1/ch1.md
@@ -21,5 +21,5 @@
 * [第五课-数据流分析理论一](https://www.bilibili.com/video/BV1A741117it)
 * [第六课-数据流分析理论二](https://www.bilibili.com/video/BV1964y1M7nL)
 
-从[这里](https://ranger-nju.gitbook.io/static-program-analysis-book/ch4)直接跳转到施工完毕区域。
+从[这里](https://ranger-nju.gitbook.io/static-program-analysis-book/ch2)直接跳转到施工完毕区域。
 
diff --git a/ch3/context-sensitivity/03-05-cs2.md b/ch3/context-sensitivity/03-05-cs2.md
@@ -47,11 +47,11 @@
 在这一部分，我们只需要理解Select的作用（对于Select的具体实现，会在后面讲解）：
 
 * ProcessCall接收两个参数，意义是：带有上下文标记的x新增一个带有上下文标记指向目标o。
-* Select接收参数（这里虽然有4个参数，但并非每种实现方式都需要用到所有的4个参数）
+* m代表目标方法。
+* Select接收参数（这里虽然有3个参数，但并非每种实现方式都需要用到所有的3个参数）
   * c。x的上下文标记
   * l。调用点本身（call site），在例子中以行号标识调用点
   * $$c' : o_i $$。receiver object
-  * m。 目标method
 * Select返回callee的context $$ c^t$$
 
 ## Context Sensitivity Variants

diff --git a/ch4/04-01-security.md b/ch4/04-01-security.md
@@ -106,7 +106,7 @@ More on Integrity-a Board Definition(在信息流安全以外的语境中，Inte
 
 More:
 
--   通过观察电量消耗、网络流量特征、缓存命中率、服务器相应时长特征，都能以某种方式获得某种程度的机密信息。
+-   通过观察电量消耗、网络流量特征、缓存命中率、服务器响应时长特征，都能以某种方式获得某种程度的机密信息。
 
 -   Side Channel: "AF缺乏淡水"
 

diff --git a/...sed-PA.assets/image-20201223184349163.png → ...sed-PA.assets/image-20201223184349163.png b/...sed-PA.assets/image-20201223184349163.png → ...sed-PA.assets/image-20201223184349163.png
diff --git a/...sed-PA.assets/image-20201223184415502.png → ...sed-PA.assets/image-20201223184415502.png b/...sed-PA.assets/image-20201223184415502.png → ...sed-PA.assets/image-20201223184415502.png
diff --git a/...sed-PA.assets/image-20201223185015690.png → ...sed-PA.assets/image-20201223185015690.png b/...sed-PA.assets/image-20201223185015690.png → ...sed-PA.assets/image-20201223185015690.png
diff --git a/...sed-PA.assets/image-20201223185231533.png → ...sed-PA.assets/image-20201223185231533.png b/...sed-PA.assets/image-20201223185231533.png → ...sed-PA.assets/image-20201223185231533.png
diff --git a/...sed-PA.assets/image-20201223185504296.png → ...sed-PA.assets/image-20201223185504296.png b/...sed-PA.assets/image-20201223185504296.png → ...sed-PA.assets/image-20201223185504296.png
diff --git a/...sed-PA.assets/image-20201223185750701.png → ...sed-PA.assets/image-20201223185750701.png b/...sed-PA.assets/image-20201223185750701.png → ...sed-PA.assets/image-20201223185750701.png
diff --git a/...sed-PA.assets/image-20201223185957740.png → ...sed-PA.assets/image-20201223185957740.png b/...sed-PA.assets/image-20201223185957740.png → ...sed-PA.assets/image-20201223185957740.png
diff --git a/...sed-PA.assets/image-20201223190539380.png → ...sed-PA.assets/image-20201223190539380.png b/...sed-PA.assets/image-20201223190539380.png → ...sed-PA.assets/image-20201223190539380.png
diff --git a/...sed-PA.assets/image-20201223190710495.png → ...sed-PA.assets/image-20201223190710495.png b/...sed-PA.assets/image-20201223190710495.png → ...sed-PA.assets/image-20201223190710495.png
diff --git a/...sed-PA.assets/image-20201223190916420.png → ...sed-PA.assets/image-20201223190916420.png b/...sed-PA.assets/image-20201223190916420.png → ...sed-PA.assets/image-20201223190916420.png
diff --git a/...sed-PA.assets/image-20201223191303096.png → ...sed-PA.assets/image-20201223191303096.png b/...sed-PA.assets/image-20201223191303096.png → ...sed-PA.assets/image-20201223191303096.png
diff --git a/...sed-PA.assets/image-20201223191451507.png → ...sed-PA.assets/image-20201223191451507.png b/...sed-PA.assets/image-20201223191451507.png → ...sed-PA.assets/image-20201223191451507.png
diff --git a/...sed-PA.assets/image-20201223191716873.png → ...sed-PA.assets/image-20201223191716873.png b/...sed-PA.assets/image-20201223191716873.png → ...sed-PA.assets/image-20201223191716873.png
diff --git a/...sed-PA.assets/image-20201223192111827.png → ...sed-PA.assets/image-20201223192111827.png b/...sed-PA.assets/image-20201223192111827.png → ...sed-PA.assets/image-20201223192111827.png
diff --git a/...sed-PA.assets/image-20201223193351919.png → ...sed-PA.assets/image-20201223193351919.png b/...sed-PA.assets/image-20201223193351919.png → ...sed-PA.assets/image-20201223193351919.png
diff --git a/...sed-PA.assets/image-20201223193808413.png → ...sed-PA.assets/image-20201223193808413.png b/...sed-PA.assets/image-20201223193808413.png → ...sed-PA.assets/image-20201223193808413.png
diff --git a/...sed-PA.assets/image-20201223194245876.png → ...sed-PA.assets/image-20201223194245876.png b/...sed-PA.assets/image-20201223194245876.png → ...sed-PA.assets/image-20201223194245876.png
diff --git a/...sed-PA.assets/image-20201223194326258.png → ...sed-PA.assets/image-20201223194326258.png b/...sed-PA.assets/image-20201223194326258.png → ...sed-PA.assets/image-20201223194326258.png
diff --git a/...sed-PA.assets/image-20201223194559347.png → ...sed-PA.assets/image-20201223194559347.png b/...sed-PA.assets/image-20201223194559347.png → ...sed-PA.assets/image-20201223194559347.png
diff --git a/...sed-PA.assets/image-20201223195004344.png → ...sed-PA.assets/image-20201223195004344.png b/...sed-PA.assets/image-20201223195004344.png → ...sed-PA.assets/image-20201223195004344.png
diff --git a/...sed-PA.assets/image-20201223195501180.png → ...sed-PA.assets/image-20201223195501180.png b/...sed-PA.assets/image-20201223195501180.png → ...sed-PA.assets/image-20201223195501180.png
diff --git a/...sed-PA.assets/image-20201223200243589.png → ...sed-PA.assets/image-20201223200243589.png b/...sed-PA.assets/image-20201223200243589.png → ...sed-PA.assets/image-20201223200243589.png
diff --git a/...sed-PA.assets/image-20201223200506410.png → ...sed-PA.assets/image-20201223200506410.png b/...sed-PA.assets/image-20201223200506410.png → ...sed-PA.assets/image-20201223200506410.png
diff --git a/...sed-PA.assets/image-20201223200617134.png → ...sed-PA.assets/image-20201223200617134.png b/...sed-PA.assets/image-20201223200617134.png → ...sed-PA.assets/image-20201223200617134.png
diff --git a/...sed-PA.assets/image-20201223200945544.png → ...sed-PA.assets/image-20201223200945544.png b/...sed-PA.assets/image-20201223200945544.png → ...sed-PA.assets/image-20201223200945544.png
diff --git a/...sed-PA.assets/image-20201223201109543.png → ...sed-PA.assets/image-20201223201109543.png b/...sed-PA.assets/image-20201223201109543.png → ...sed-PA.assets/image-20201223201109543.png
diff --git a/...sed-PA.assets/image-20201223201354991.png → ...sed-PA.assets/image-20201223201354991.png b/...sed-PA.assets/image-20201223201354991.png → ...sed-PA.assets/image-20201223201354991.png
diff --git a/...sed-PA.assets/image-20201223201428109.png → ...sed-PA.assets/image-20201223201428109.png b/...sed-PA.assets/image-20201223201428109.png → ...sed-PA.assets/image-20201223201428109.png
diff --git a/...sed-PA.assets/image-20201223201612811.png → ...sed-PA.assets/image-20201223201612811.png b/...sed-PA.assets/image-20201223201612811.png → ...sed-PA.assets/image-20201223201612811.png
diff --git a/...sed-PA.assets/image-20201223201746117.png → ...sed-PA.assets/image-20201223201746117.png b/...sed-PA.assets/image-20201223201746117.png → ...sed-PA.assets/image-20201223201746117.png
diff --git a/...sed-PA.assets/image-20201223201956852.png → ...sed-PA.assets/image-20201223201956852.png b/...sed-PA.assets/image-20201223201956852.png → ...sed-PA.assets/image-20201223201956852.png
diff --git a/...sed-PA.assets/image-20201223202140289.png → ...sed-PA.assets/image-20201223202140289.png b/...sed-PA.assets/image-20201223202140289.png → ...sed-PA.assets/image-20201223202140289.png
diff --git a/ch5/05-01-Datalog-Based-PA.md → ch4/04-02-Datalog-Based-PA.md b/ch5/05-01-Datalog-Based-PA.md → ch4/04-02-Datalog-Based-PA.md
@@ -13,11 +13,11 @@ Datalog是一种命令式（Declarative）的编程语言。
 
 如果用Imperative的编程方式做指针分析，很麻烦。
 
-<img src="05-01-Datalog-Based-PA.assets/image-20201223184349163.png" style="zoom:50%;" />
+<img src="04-02-Datalog-Based-PA.assets/image-20201223184349163.png" style="zoom:50%;" />
 
 而如果用Declarative的方式做编程分析，能够极大地简化实现。
 
-<img src="05-01-Datalog-Based-PA.assets/image-20201223184415502.png" style="zoom:50%;" />
+<img src="04-02-Datalog-Based-PA.assets/image-20201223184415502.png" style="zoom:50%;" />
 
 # Introduction to Datalog
 
@@ -36,18 +36,18 @@ Datalog是一种命令式（Declarative）的编程语言。
 
 谓词(Predicates)是datalog中的一个主要组成部分，可以看作是数据所组成的一个表(table of data)，每一行都代表一个事实(fact)。例如：
 
-<img src="05-01-Datalog-Based-PA.assets/image-20201223185015690.png" style="zoom:50%;" />
+<img src="04-02-Datalog-Based-PA.assets/image-20201223185015690.png" style="zoom:50%;" />
 
 ### Atoms
 
 原子(Atoms)是Datalog中的基本元素，组成和例子如下：
 
-<img src="05-01-Datalog-Based-PA.assets/image-20201223185231533.png" style="zoom:50%;" />
+<img src="04-02-Datalog-Based-PA.assets/image-20201223185231533.png" style="zoom:50%;" />
 
 Atoms可以分成两类
 
 -   Relational Atoms
-    -   <img src="05-01-Datalog-Based-PA.assets/image-20201223185504296.png" style="zoom:50%;" />
+    -   <img src="04-02-Datalog-Based-PA.assets/image-20201223185504296.png" style="zoom:50%;" />
 -   Arithmetic Atoms
     -   如`age >= 18`
 
@@ -57,17 +57,17 @@ Atoms可以分成两类
 
 Datalog使用规则来进行推导(inference)，其定义如下：
 
-<img src="05-01-Datalog-Based-PA.assets/image-20201223185750701.png" style="zoom:50%;" />
+<img src="04-02-Datalog-Based-PA.assets/image-20201223185750701.png" style="zoom:50%;" />
 
 当Body中的所有表达式都为True时，Head才为True，如：
 
-<img src="05-01-Datalog-Based-PA.assets/image-20201223185957740.png" style="zoom:50%;" />
+<img src="04-02-Datalog-Based-PA.assets/image-20201223185957740.png" style="zoom:50%;" />
 
 求解过程(Interpretation of Datalog Rules)——枚举Body中所有关系表达式的可能取值组合，进而得到新的predicate/table。例如：
 
-<img src="05-01-Datalog-Based-PA.assets/image-20201223190539380.png" style="zoom:50%;" />
+<img src="04-02-Datalog-Based-PA.assets/image-20201223190539380.png" style="zoom:50%;" />
 
-<img src="05-01-Datalog-Based-PA.assets/image-20201223190710495.png" style="zoom:50%;" />
+<img src="04-02-Datalog-Based-PA.assets/image-20201223190710495.png" style="zoom:50%;" />
 
 谓词分为两类：EDB & IDB。
 
@@ -78,23 +78,23 @@ Datalog使用规则来进行推导(inference)，其定义如下：
 
 例如：
 
-<img src="05-01-Datalog-Based-PA.assets/image-20201223190916420.png" style="zoom:50%;" />
+<img src="04-02-Datalog-Based-PA.assets/image-20201223190916420.png" style="zoom:50%;" />
 
 ### Logic Or
 
 以上例子实际上是逻辑与，而逻辑或则有两种实现方式：
 
-<img src="05-01-Datalog-Based-PA.assets/image-20201223191303096.png" style="zoom:50%;" />
+<img src="04-02-Datalog-Based-PA.assets/image-20201223191303096.png" style="zoom:50%;" />
 
 此外还需要考虑运算优先级的问题，建议在书写程序时用括号明确地标识期望的运算优先级：`H<-A,(B;C)`。
 
 ### Logic Not/Negation
 
-<img src="05-01-Datalog-Based-PA.assets/image-20201223191451507.png" style="zoom:50%;" />
+<img src="04-02-Datalog-Based-PA.assets/image-20201223191451507.png" style="zoom:50%;" />
 
 ### Recursion
 
-<img src="05-01-Datalog-Based-PA.assets/image-20201223191716873.png" style="zoom:50%;" />
+<img src="04-02-Datalog-Based-PA.assets/image-20201223191716873.png" style="zoom:50%;" />
 
 ## Rule Satety
 
@@ -103,7 +103,7 @@ Datalog使用规则来进行推导(inference)，其定义如下：
 -   `A(x) <- B(y), x > y.`
 -   `A(x) <- B(y), !C(x,y).`
 
-<img src="05-01-Datalog-Based-PA.assets/image-20201223192111827.png" style="zoom:50%;" />
+<img src="04-02-Datalog-Based-PA.assets/image-20201223192111827.png" style="zoom:50%;" />
 
 ~~第一次学看不出问题也没问题的~~
 
@@ -119,13 +119,13 @@ Datalog使用规则来进行推导(inference)，其定义如下：
 
 `A(x) <- B(x), !A(x)`
 
-<img src="05-01-Datalog-Based-PA.assets/image-20201223193351919.png" style="zoom:50%;" />
+<img src="04-02-Datalog-Based-PA.assets/image-20201223193351919.png" style="zoom:50%;" />
 
 对应地有第二个准则：**不要把recursion和negation写在同一条规则里**，*即避免写出非A推导出A这样的规则*。
 
 ## Execution of Datalog Program
 
-<img src="05-01-Datalog-Based-PA.assets/image-20201223193808413.png" style="zoom:50%;" />
+<img src="04-02-Datalog-Based-PA.assets/image-20201223193808413.png" style="zoom:50%;" />
 
 Datalog的两大重要特性：
 
@@ -142,25 +142,25 @@ Datalog的两大重要特性：
 -   IDB：指针分析的结果
 -   Rules：指针分析的规则
 
-<img src="05-01-Datalog-Based-PA.assets/image-20201223194245876.png" style="zoom:50%;" />
+<img src="04-02-Datalog-Based-PA.assets/image-20201223194245876.png" style="zoom:50%;" />
 
 和之前一样，我们把Call放到最后处理。
 
 ## Datalog Model-EDB&IDB
 
 我们首先需要对前四条语句建模。输入的EDB代表了4个存储相应类型语句的table，输出为Variable和Field的指向关系。
 
-<img src="05-01-Datalog-Based-PA.assets/image-20201223194326258.png" style="zoom:50%;" />
+<img src="04-02-Datalog-Based-PA.assets/image-20201223194326258.png" style="zoom:50%;" />
 
 一个关于EDB的例子如下：
 
-<img src="05-01-Datalog-Based-PA.assets/image-20201223194559347.png" style="zoom:50%;" />
+<img src="04-02-Datalog-Based-PA.assets/image-20201223194559347.png" style="zoom:50%;" />
 
 ## Datalog Rules
 
 Body按照红线所示代表前提，Head按照蓝线所示代表结论。
 
-<img src="05-01-Datalog-Based-PA.assets/image-20201223195004344.png" style="zoom:50%;" />
+<img src="04-02-Datalog-Based-PA.assets/image-20201223195004344.png" style="zoom:50%;" />
 
 ## Example 
 
@@ -178,71 +178,71 @@ e = d.f;
 
 结果如下：
 
-<img src="05-01-Datalog-Based-PA.assets/image-20201223195501180.png" style="zoom:50%;" />
+<img src="04-02-Datalog-Based-PA.assets/image-20201223195501180.png" style="zoom:50%;" />
 
 ## PA with Calls
 
 回顾指针分析中Call的规则：
 
-<img src="05-01-Datalog-Based-PA.assets/image-20201223200506410.png" style="zoom:50%;" />
+<img src="04-02-Datalog-Based-PA.assets/image-20201223200506410.png" style="zoom:50%;" />
 
 ### This
 
 首先，我们需要引入新的EDB和IDB：
 
 （VCall即Virtual Call，ThisVar即This Variable）
 
-<img src="05-01-Datalog-Based-PA.assets/image-20201223200243589.png" style="zoom:50%;" />
+<img src="04-02-Datalog-Based-PA.assets/image-20201223200243589.png" style="zoom:50%;" />
 
 我们能够同时知道三个信息：
 
 1.  this指向对象o
 2.  方法m是可达的
 3.  方法m可达是因为在l行处存在对方法m的调用
 
-<img src="05-01-Datalog-Based-PA.assets/image-20201223200617134.png" style="zoom:50%;" />
+<img src="04-02-Datalog-Based-PA.assets/image-20201223200617134.png" style="zoom:50%;" />
 
 ### Parameters
 
 接下来要处理参数的传递，与之前类似，引入EDB标识Argument(调用语句行号，参数标号和参数本身)和Parameter(被调用方法，参数标号和参数本身)：
 
-<img src="05-01-Datalog-Based-PA.assets/image-20201223200945544.png" style="zoom:50%;" />
+<img src="04-02-Datalog-Based-PA.assets/image-20201223200945544.png" style="zoom:50%;" />
 
 而对应用Datalog书写的规则如果用自然语言描述，就是处理行号l处对m的调用时，根据形参和实参的信息，将实参已经有的指向关系传递给形参数。
 
-<img src="05-01-Datalog-Based-PA.assets/image-20201223201109543.png" style="zoom:50%;" />
+<img src="04-02-Datalog-Based-PA.assets/image-20201223201109543.png" style="zoom:50%;" />
 
 ### Return Value
 
 引入EDB：
 
-<img src="05-01-Datalog-Based-PA.assets/image-20201223201354991.png" style="zoom:50%;" />
+<img src="04-02-Datalog-Based-PA.assets/image-20201223201354991.png" style="zoom:50%;" />
 
 对应的Datalog Rule，处理返回值的指向关系：
 
-<img src="05-01-Datalog-Based-PA.assets/image-20201223201428109.png" style="zoom:50%;" />
+<img src="04-02-Datalog-Based-PA.assets/image-20201223201428109.png" style="zoom:50%;" />
 
 ### Sum up
 
 以上三个部分总结起来，就能得到以下的Datalog Rules。
 
-<img src="05-01-Datalog-Based-PA.assets/image-20201223201612811.png" style="zoom:50%;" />
+<img src="04-02-Datalog-Based-PA.assets/image-20201223201612811.png" style="zoom:50%;" />
 
 进而能得到全程序的分析算法如下。值得一提的是，在VarPointsTo规则中，添加Reachable(m)跳过不可达方法中的对象。而其他规则不需要加这一条件，则是因为它们都有VarPointsTo规则作为Body的一部分。
 
-<img src="05-01-Datalog-Based-PA.assets/image-20201223201746117.png" style="zoom:50%;" />
+<img src="04-02-Datalog-Based-PA.assets/image-20201223201746117.png" style="zoom:50%;" />
 
 # Taint Analysis via Datalog
 
 ## Datalog Model
 
-同样需要用户提供Source和Sink。输出被标记的数据可能流到的Sink方法。<img src="05-01-Datalog-Based-PA.assets/image-20201223201956852.png" style="zoom:50%;" />
+同样需要用户提供Source和Sink。输出被标记的数据可能流到的Sink方法。<img src="04-02-Datalog-Based-PA.assets/image-20201223201956852.png" style="zoom:50%;" />
 
 ## Datalog Rules
 
 说明：这里参数列表中用`—`表示通配符，即不关心枚举时这个位置取什么值。
 
-<img src="05-01-Datalog-Based-PA.assets/image-20201223202140289.png" style="zoom:50%;" />
+<img src="04-02-Datalog-Based-PA.assets/image-20201223202140289.png" style="zoom:50%;" />
 
 # Key Points
 

diff --git a/ch4/ch4.md b/ch4/ch4.md
@@ -1,3 +1,6 @@
-# 指针分析应用
+# 指针分析应用、实现与其他
 
-接下来说说基于指针分析的污点分析(Taint Analysis)。
+接下来说说基于指针分析的静态分析相关技术知识：
+
+-   安全应用方向的污点分析(Taint Analysis)。
+-   如何用声明式的语言Datalog实现指针分析。
diff --git a/ch5/05-01-IFDS.assets/image-20201224200732868.png b/ch5/05-01-IFDS.assets/image-20201224200732868.png
diff --git a/ch5/05-01-IFDS.assets/image-20201224201312394.png b/ch5/05-01-IFDS.assets/image-20201224201312394.png
diff --git a/ch5/05-01-IFDS.assets/image-20201224201921134.png b/ch5/05-01-IFDS.assets/image-20201224201921134.png
diff --git a/ch5/05-01-IFDS.assets/image-20201231184204252.png b/ch5/05-01-IFDS.assets/image-20201231184204252.png
diff --git a/ch5/05-01-IFDS.assets/image-20201231184249491.png b/ch5/05-01-IFDS.assets/image-20201231184249491.png
diff --git a/ch5/05-01-IFDS.assets/image-20201231185030483.png b/ch5/05-01-IFDS.assets/image-20201231185030483.png
diff --git a/ch5/05-01-IFDS.assets/image-20201231185419312.png b/ch5/05-01-IFDS.assets/image-20201231185419312.png
diff --git a/ch5/05-01-IFDS.assets/image-20201231185639039.png b/ch5/05-01-IFDS.assets/image-20201231185639039.png
diff --git a/ch5/05-01-IFDS.assets/image-20201231190743854.png b/ch5/05-01-IFDS.assets/image-20201231190743854.png
diff --git a/ch5/05-01-IFDS.assets/image-20201231191027738.png b/ch5/05-01-IFDS.assets/image-20201231191027738.png
diff --git a/ch5/05-01-IFDS.assets/image-20201231191043066.png b/ch5/05-01-IFDS.assets/image-20201231191043066.png
diff --git a/ch5/05-01-IFDS.assets/image-20201231191553161.png b/ch5/05-01-IFDS.assets/image-20201231191553161.png
diff --git a/ch5/05-01-IFDS.assets/image-20201231191923368.png b/ch5/05-01-IFDS.assets/image-20201231191923368.png
diff --git a/ch5/05-01-IFDS.assets/image-20201231192128228.png b/ch5/05-01-IFDS.assets/image-20201231192128228.png
diff --git a/ch5/05-01-IFDS.assets/image-20201231193305999.png b/ch5/05-01-IFDS.assets/image-20201231193305999.png
diff --git a/ch5/05-01-IFDS.assets/image-20201231193404292.png b/ch5/05-01-IFDS.assets/image-20201231193404292.png
diff --git a/ch5/05-01-IFDS.assets/image-20201231193423998.png b/ch5/05-01-IFDS.assets/image-20201231193423998.png
diff --git a/ch5/05-01-IFDS.assets/image-20201231193951460.png b/ch5/05-01-IFDS.assets/image-20201231193951460.png
diff --git a/ch5/05-01-IFDS.assets/image-20201231194039123.png b/ch5/05-01-IFDS.assets/image-20201231194039123.png
diff --git a/ch5/05-01-IFDS.assets/image-20201231194247160.png b/ch5/05-01-IFDS.assets/image-20201231194247160.png
diff --git a/ch5/05-01-IFDS.assets/image-20201231194752369.png b/ch5/05-01-IFDS.assets/image-20201231194752369.png
diff --git a/ch5/05-01-IFDS.assets/image-20201231195241378.png b/ch5/05-01-IFDS.assets/image-20201231195241378.png
diff --git a/ch5/05-01-IFDS.assets/image-20201231195259204.png b/ch5/05-01-IFDS.assets/image-20201231195259204.png
diff --git a/ch5/05-01-IFDS.assets/image-20201231195711805.png b/ch5/05-01-IFDS.assets/image-20201231195711805.png
diff --git a/ch5/05-01-IFDS.assets/image-20201231195848429.png b/ch5/05-01-IFDS.assets/image-20201231195848429.png
diff --git a/ch5/05-01-IFDS.assets/image-20201231200145777.png b/ch5/05-01-IFDS.assets/image-20201231200145777.png
diff --git a/ch5/05-01-IFDS.assets/image-20201231200214568.png b/ch5/05-01-IFDS.assets/image-20201231200214568.png
diff --git a/ch5/05-01-IFDS.assets/image-20201231200351183.png b/ch5/05-01-IFDS.assets/image-20201231200351183.png
diff --git a/ch5/05-01-IFDS.assets/image-20201231200406642.png b/ch5/05-01-IFDS.assets/image-20201231200406642.png
diff --git a/ch5/05-01-IFDS.assets/image-20201231200623730.png b/ch5/05-01-IFDS.assets/image-20201231200623730.png
diff --git a/ch5/05-01-IFDS.assets/image-20201231200656200.png b/ch5/05-01-IFDS.assets/image-20201231200656200.png
diff --git a/ch5/05-01-IFDS.assets/image-20201231201028380.png b/ch5/05-01-IFDS.assets/image-20201231201028380.png
diff --git a/ch5/05-01-IFDS.assets/image-20201231202231523.png b/ch5/05-01-IFDS.assets/image-20201231202231523.png
diff --git a/ch5/05-01-IFDS.assets/image-20201231202450969.png b/ch5/05-01-IFDS.assets/image-20201231202450969.png
-Original file line number
+Diff line change
@@ Expand Up @@
     More:
-    -   通过观察电量消耗、网络流量特征、缓存命中率、服务器相应时长特征，都能以某种方式获得某种程度的机密信息。
+    -   通过观察电量消耗、网络流量特征、缓存命中率、服务器响应时长特征，都能以某种方式获得某种程度的机密信息。
     -   Side Channel: "AF缺乏淡水"
@@ Expand Down @@