diff --git a/notes/7.md b/notes/7.md
index 2ff412a..387e507 100644
--- a/notes/7.md
+++ b/notes/7.md
@@ -17,12 +17,10 @@ Keep the top of the stack in a register, so `add` requires only a single memory
 * `pop`
 * `add`
 
-
 ### Code Generation From Stack Machine
 
 Assume that stack grows towards lower addresses.
 
-
 ## MIPS
 
 32 regs
@@ -36,7 +34,6 @@ Assume that stack grows towards lower addresses.
 * `li`
 * `mv`
 
-
 Converting Stack to MIPS ISA
 
 * `acc <- i`
@@ -44,13 +41,56 @@ Converting Stack to MIPS ISA
 
 ### Optimizing
 
-
+no
 
 ## Branch
 
 `beq $1 $2 lbl`
 `b lbl`
 
-
 ## Function
 
+At Caller Side:
+1. saves the `$fp` to the stack
+2. saves the actual params in reverse order
+3. saves the return address to `$ra`
+
+At Callee Side:
+1. set `$fp` to `$sp`
+2. the callee saves the return address
+3. ...
+
+## Temp Var
+
+Many various intermediate vars should be stored in the AR. But compiler can statically know how many temporary variables there are.
+
+Let $NT(e)$ is defined recursively by:
+$$NT(e1 + e2) = \max (NT(e1), NT(e2) + 1)$$
+
+for example:
+
+### CG using NT
+
+add new args to the `cgen(e, nt)`
+
+reduce number of decrease `$sp` (`addi $sp $sp -4`)
+
+
+## Intermediate Representation (IR)
+
+Before:
+Each Languages need respective, different Optimizer.
+
+Now:
+Common IR optimizer.
+
+
+IR is language-independent and machine-independent optimization.
+
+### High-Level Assembly
+
+It uses unlimited number of registers.
+It uses assembly-like control structures (jmp and lbl).
+opcodes but some are higher level
+
+igen(e, t)
\ No newline at end of file
diff --git a/notes/8.md b/notes/8.md
new file mode 100644
index 0000000..c82b0b1
--- /dev/null
+++ b/notes/8.md
@@ -0,0 +1,52 @@
+# Dataflow Analysis
+
+Optimization means improving resource utilization not changing what the program computes.
+Resource utilization means many things:
+* **Execution time**
+* Code size
+* Network messages sent.
+
+## Basic Block (BB)
+
+A BB is a maximum sequence of instructions with **no labels**, **no jumps**
+All instructions in a BB has fixed control flow.
+
+## Control Flow Graph
+
+```mermaid
+flowchart TD
+    Entry --> A
+    A[BB1] --> B[BB2]
+    A --> C[BB3]
+    B --> D[BB4]
+    C --> D
+    D --> E[BB5]
+    E --> G[BB7]
+    E --> F[BB6]
+    G --> Exit
+    F --> Exit
+```
+
+## Local Optimization
+
+### Algebraic Simplification
+
+x := x + 0 -> x := 0
+y := y ** 2 -> y := y * y
+x := x * 8 -> x := x << 3
+x := x * 15 -> t := x << 4; x := t - x
+
+### Constant Folding
+
+x := 2 + 2 -> x := 4
+if 2 < 0 jump L -> nop
+if 2 > 0 jump L -> jump L
+
+But Constant folding can be dangerous on cross-compilation (in precision).
+
+
+### Unreachable Code
+
+### Dead Code Elimination
+
+## Global Optimization
\ No newline at end of file
diff --git a/pdf/L8.pdf b/pdf/L8.pdf
new file mode 100644
index 0000000..7611305
Binary files /dev/null and b/pdf/L8.pdf differ