summaryrefslogtreecommitdiff
path: root/Content/posts
diff options
context:
space:
mode:
Diffstat (limited to 'Content/posts')
-rw-r--r--Content/posts/2023-10-04-bomb-lab.md1047
-rw-r--r--Content/posts/2023-10-05-attack-lab.md22
2 files changed, 1069 insertions, 0 deletions
diff --git a/Content/posts/2023-10-04-bomb-lab.md b/Content/posts/2023-10-04-bomb-lab.md
new file mode 100644
index 0000000..c805279
--- /dev/null
+++ b/Content/posts/2023-10-04-bomb-lab.md
@@ -0,0 +1,1047 @@
+---
+date: 2023-10-04 13:12
+description: Walkthrough of Phases 1-6 of Bomb Lab for CSCI 2400 Computer Systems Lab 2
+tags: gdb, reverse-engineering, c++, csci2400, assembly
+---
+
+# Bomb Lab
+
+## Introduction
+
+Lab 2 for CSCI 2400 @ CU Boulder - Computer Systems
+
+> The nefarious Dr. Evil has planted a slew of “binary bombs” on our class machines. A binary bomb is a program that consists of a sequence of phases. Each phase expects you to type a particular string on stdin. If you type the correct string, then the phase is defused and the bomb proceeds to the next phase. Otherwise, the bomb explodes by printing "BOOM!!!" and then terminating. The bomb is defused when every phase has been defused.
+
+> There are too many bombs for us to deal with, so we are giving each student a bomb to defuse. Your mission, which you have no choice but to accept, is to defuse your bomb before the due date. Good luck, and welcome to the bomb squad!
+
+I like using objdump to disassemble the code and get a broad overview of what is happening before I start.
+
+`objdump -d bomb > dis.txt`
+
+*Note: I am not sure about the history of the bomb lab. I think it started at CMU.*
+
+## Phase 1
+
+```shell
+joxxxn@jupyter-nxxh6xx8:~/lab2-bomblab-navanchauhan/bombbomb$ gdb -ex 'break phase_1' -ex 'break explode_bomb' -ex 'run' ./bomb
+GNU gdb (Ubuntu 12.1-0ubuntu1~22.04) 12.1
+Copyright (C) 2022 Free Software Foundation, Inc.
+License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html>
+This is free software: you are free to change and redistribute it.
+There is NO WARRANTY, to the extent permitted by law.
+Type "show copying" and "show warranty" for details.
+This GDB was configured as "x86_64-linux-gnu".
+Type "show configuration" for configuration details.
+For bug reporting instructions, please see:
+<https://www.gnu.org/software/gdb/bugs/>.
+Find the GDB manual and other documentation resources online at:
+ <http://www.gnu.org/software/gdb/documentation/>.
+
+For help, type "help".
+Type "apropos word" to search for commands related to "word"...
+Reading symbols from ./bomb...
+Breakpoint 1 at 0x15c7
+Breakpoint 2 at 0x1d4a
+Starting program: /home/joxxxn/lab2-bomblab-navanchauhan/bombbomb/bomb
+[Thread debugging using libthread_db enabled]
+Using host libthread_db library "/lib/x86_64-linux-gnu/libthread_db.so.1".
+Welcome to my fiendish little bomb. You have 6 phases with
+which to blow yourself up. Have a nice day!
+test string
+
+Breakpoint 1, 0x00005555555555c7 in phase_1 ()
+(gdb) dias phase_1
+Undefined command: "dias". Try "help".
+(gdb) disas phase_1
+Dump of assembler code for function phase_1:
+=> 0x00005555555555c7 <+0>: endbr64
+ 0x00005555555555cb <+4>: sub $0x8,%rsp
+ 0x00005555555555cf <+8>: lea 0x1b7a(%rip),%rsi # 0x555555557150
+ 0x00005555555555d6 <+15>: call 0x555555555b31 <strings_not_equal>
+ 0x00005555555555db <+20>: test %eax,%eax
+ 0x00005555555555dd <+22>: jne 0x5555555555e4 <phase_1+29>
+ 0x00005555555555df <+24>: add $0x8,%rsp
+ 0x00005555555555e3 <+28>: ret
+ 0x00005555555555e4 <+29>: call 0x555555555d4a <explode_bomb>
+ 0x00005555555555e9 <+34>: jmp 0x5555555555df <phase_1+24>
+End of assembler dump.
+(gdb) print 0x555555557150
+$1 = 93824992244048
+(gdb) x/1s 0x555555557150
+0x555555557150: "Controlling complexity is the essence of computer programming."
+(gdb)
+```
+
+## Phase 2
+
+```shell
+Phase 1 defused. How about the next one?
+1 2 3 4 5 6
+
+Breakpoint 1, 0x00005555555555eb in phase_2 ()
+(gdb) disas
+Dump of assembler code for function phase_2:
+=> 0x00005555555555eb <+0>: endbr64
+ 0x00005555555555ef <+4>: push %rbp
+ 0x00005555555555f0 <+5>: push %rbx
+ 0x00005555555555f1 <+6>: sub $0x28,%rsp
+ 0x00005555555555f5 <+10>: mov %rsp,%rsi
+ 0x00005555555555f8 <+13>: call 0x555555555d97 <read_six_numbers>
+ 0x00005555555555fd <+18>: cmpl $0x0,(%rsp)
+ 0x0000555555555601 <+22>: js 0x55555555560d <phase_2+34>
+ 0x0000555555555603 <+24>: mov %rsp,%rbp
+ 0x0000555555555606 <+27>: mov $0x1,%ebx
+ 0x000055555555560b <+32>: jmp 0x555555555620 <phase_2+53>
+ 0x000055555555560d <+34>: call 0x555555555d4a <explode_bomb>
+ 0x0000555555555612 <+39>: jmp 0x555555555603 <phase_2+24>
+ 0x0000555555555614 <+41>: add $0x1,%ebx
+ 0x0000555555555617 <+44>: add $0x4,%rbp
+ 0x000055555555561b <+48>: cmp $0x6,%ebx
+ 0x000055555555561e <+51>: je 0x555555555631 <phase_2+70>
+ 0x0000555555555620 <+53>: mov %ebx,%eax
+ 0x0000555555555622 <+55>: add 0x0(%rbp),%eax
+ 0x0000555555555625 <+58>: cmp %eax,0x4(%rbp)
+ 0x0000555555555628 <+61>: je 0x555555555614 <phase_2+41>
+ 0x000055555555562a <+63>: call 0x555555555d4a <explode_bomb>
+ 0x000055555555562f <+68>: jmp 0x555555555614 <phase_2+41>
+ 0x0000555555555631 <+70>: add $0x28,%rsp
+ 0x0000555555555635 <+74>: pop %rbx
+ 0x0000555555555636 <+75>: pop %rbp
+ 0x0000555555555637 <+76>: ret
+End of assembler dump.
+(gdb)
+```
+
+```shell
+ 0x00005555555555fd <+18>: cmpl $0x0,(%rsp)
+ 0x0000555555555601 <+22>: js 0x55555555560d <phase_2+34>
+...
+ 0x000055555555560d <+34>: call 0x555555555d4a <explode_bomb>
+```
+
+The program first compares if the first number is not 0. If the number is not 0, then the `cmpl` instruction returns a negative value. The `js` instruction stands for jump if sign -> causing a jump to the specified address if the sign bit is set. This would result in the explode_bomb function being called.
+
+
+```shell
+ 0x0000555555555603 <+24>: mov %rsp,%rbp
+ 0x0000555555555606 <+27>: mov $0x1,%ebx
+```
+
+`%rsp` in x86-64 asm, is the stack pointer i.e. it points to the top of the current stack frame. Since the program just read six numbers, the top of the stack (`%rsp`) contains the address of the first number.
+
+
+By executing `mov %rsp,%rbp` we are setting the base pointer (`%rbp`) to point to this address.
+
+Now, for the second instruction `mov $0x1,%ebx`, we are initalising the `%ebx` register with the value 1. Based on the assembly code, you can see that this is being used as a counter/index for the loop.
+
+
+```shell
+ 0x000055555555560b <+32>: jmp 0x555555555620 <phase_2+53>
+```
+
+The program now jumps to <phase_2+53>
+
+```shell
+ 0x0000555555555620 <+53>: mov %ebx,%eax
+ 0x0000555555555622 <+55>: add 0x0(%rbp),%eax
+ 0x0000555555555625 <+58>: cmp %eax,0x4(%rbp)
+ 0x0000555555555628 <+61>: je 0x555555555614 <phase_2+41>
+```
+
+Here, the value from `%ebx` is copied to the `%eax` register. For this iteration, the value should be 1.
+
+Then, the value at the memory location pointed by `%rbp` is added to the value in `%eax`. For now, 0 is added (the first number that we read).
+
+`cmp %eax,0x4(%rbp)` - The instruction compares the value in %eax to the value at the memory address `%rbp + 4`. Since Integers in this context are stored using a word of memory of 4 bytes, this indicates it checks against the second number in the sequence.
+
+`je 0x555555555614 <phase_2+41>` - The program will jump to `phase_2+41` if the previous `cmp` instruction determined the values as equal.
+
+```shell
+ 0x0000555555555614 <+41>: add $0x1,%ebx
+ 0x0000555555555617 <+44>: add $0x4,%rbp
+ 0x000055555555561b <+48>: cmp $0x6,%ebx
+ 0x000055555555561e <+51>: je 0x555555555631 <phase_2+70>
+ 0x0000555555555620 <+53>: mov %ebx,%eax
+ 0x0000555555555622 <+55>: add 0x0(%rbp),%eax
+ 0x0000555555555625 <+58>: cmp %eax,0x4(%rbp)
+ 0x0000555555555628 <+61>: je 0x555555555614 <phase_2+41>
+```
+
+Here, we can see that the program increments `%ebx` by 1, adds a 4 byte offset to `%rbp` (the number we will be matching now), and checks if `%ebx` is equal to 6. If it is, it breaks the loop and jumps to `<phase_2+70>` succesfully finishing this stage.
+
+Now, given that we know the first two numbers in the sequence are `0 1`, we can calculate the other numbers by following the pattern of adding the counter and the value of the previous number.
+
+Thus,
+
+* 3rd number = 1 (previous value) + 2 = 3
+* 4th number = 3 (prev value) + 3 = 6
+* 5th number = 6 (prev value) + 4 = 10
+* 6th number = 10 (prev value) + 5 = 15
+
+
+```shell
+...
+Phase 1 defused. How about the next one?
+0 1 3 6 10 15
+
+Breakpoint 1, 0x00005555555555eb in phase_2 ()
+(gdb) continue
+Continuing.
+That's number 2. Keep going!
+```
+
+## Phase 3
+
+Let us look at the disassembled code first
+
+```shell
+0000000000001638 <phase_3>:
+ 1638: f3 0f 1e fa endbr64
+ 163c: 48 83 ec 18 sub $0x18,%rsp
+ 1640: 48 8d 4c 24 07 lea 0x7(%rsp),%rcx
+ 1645: 48 8d 54 24 0c lea 0xc(%rsp),%rdx
+ 164a: 4c 8d 44 24 08 lea 0x8(%rsp),%r8
+ 164f: 48 8d 35 60 1b 00 00 lea 0x1b60(%rip),%rsi # 31b6 <_IO_stdin_used+0x1b6>
+ 1656: b8 00 00 00 00 mov $0x0,%eax
+ 165b: e8 80 fc ff ff call 12e0 <__isoc99_sscanf@plt>
+ 1660: 83 f8 02 cmp $0x2,%eax
+ 1663: 7e 20 jle 1685 <phase_3+0x4d>
+ 1665: 83 7c 24 0c 07 cmpl $0x7,0xc(%rsp)
+ 166a: 0f 87 0d 01 00 00 ja 177d <phase_3+0x145>
+ 1670: 8b 44 24 0c mov 0xc(%rsp),%eax
+ 1674: 48 8d 15 55 1b 00 00 lea 0x1b55(%rip),%rdx # 31d0 <_IO_stdin_used+0x1d0>
+ 167b: 48 63 04 82 movslq (%rdx,%rax,4),%rax
+ 167f: 48 01 d0 add %rdx,%rax
+ 1682: 3e ff e0 notrack jmp *%rax
+ 1685: e8 c0 06 00 00 call 1d4a <explode_bomb>
+ 168a: eb d9 jmp 1665 <phase_3+0x2d>
+ 168c: b8 63 00 00 00 mov $0x63,%eax
+ 1691: 81 7c 24 08 3d 02 00 cmpl $0x23d,0x8(%rsp)
+ 1698: 00
+ 1699: 0f 84 e8 00 00 00 je 1787 <phase_3+0x14f>
+ 169f: e8 a6 06 00 00 call 1d4a <explode_bomb>
+ 16a4: b8 63 00 00 00 mov $0x63,%eax
+ 16a9: e9 d9 00 00 00 jmp 1787 <phase_3+0x14f>
+ 16ae: b8 61 00 00 00 mov $0x61,%eax
+ 16b3: 81 7c 24 08 27 01 00 cmpl $0x127,0x8(%rsp)
+ 16ba: 00
+ 16bb: 0f 84 c6 00 00 00 je 1787 <phase_3+0x14f>
+ 16c1: e8 84 06 00 00 call 1d4a <explode_bomb>
+ 16c6: b8 61 00 00 00 mov $0x61,%eax
+ 16cb: e9 b7 00 00 00 jmp 1787 <phase_3+0x14f>
+ 16d0: b8 78 00 00 00 mov $0x78,%eax
+ 16d5: 81 7c 24 08 e7 02 00 cmpl $0x2e7,0x8(%rsp)
+ 16dc: 00
+ 16dd: 0f 84 a4 00 00 00 je 1787 <phase_3+0x14f>
+ 16e3: e8 62 06 00 00 call 1d4a <explode_bomb>
+ 16e8: b8 78 00 00 00 mov $0x78,%eax
+ 16ed: e9 95 00 00 00 jmp 1787 <phase_3+0x14f>
+ 16f2: b8 64 00 00 00 mov $0x64,%eax
+ 16f7: 81 7c 24 08 80 02 00 cmpl $0x280,0x8(%rsp)
+ 16fe: 00
+ 16ff: 0f 84 82 00 00 00 je 1787 <phase_3+0x14f>
+ 1705: e8 40 06 00 00 call 1d4a <explode_bomb>
+ 170a: b8 64 00 00 00 mov $0x64,%eax
+ 170f: eb 76 jmp 1787 <phase_3+0x14f>
+ 1711: b8 6d 00 00 00 mov $0x6d,%eax
+ 1716: 81 7c 24 08 ff 02 00 cmpl $0x2ff,0x8(%rsp)
+ 171d: 00
+ 171e: 74 67 je 1787 <phase_3+0x14f>
+ 1720: e8 25 06 00 00 call 1d4a <explode_bomb>
+ 1725: b8 6d 00 00 00 mov $0x6d,%eax
+ 172a: eb 5b jmp 1787 <phase_3+0x14f>
+ 172c: b8 71 00 00 00 mov $0x71,%eax
+ 1731: 81 7c 24 08 75 03 00 cmpl $0x375,0x8(%rsp)
+ 1738: 00
+ 1739: 74 4c je 1787 <phase_3+0x14f>
+ 173b: e8 0a 06 00 00 call 1d4a <explode_bomb>
+ 1740: b8 71 00 00 00 mov $0x71,%eax
+ 1745: eb 40 jmp 1787 <phase_3+0x14f>
+ 1747: b8 79 00 00 00 mov $0x79,%eax
+ 174c: 81 7c 24 08 94 02 00 cmpl $0x294,0x8(%rsp)
+ 1753: 00
+ 1754: 74 31 je 1787 <phase_3+0x14f>
+ 1756: e8 ef 05 00 00 call 1d4a <explode_bomb>
+ 175b: b8 79 00 00 00 mov $0x79,%eax
+ 1760: eb 25 jmp 1787 <phase_3+0x14f>
+ 1762: b8 79 00 00 00 mov $0x79,%eax
+ 1767: 81 7c 24 08 88 02 00 cmpl $0x288,0x8(%rsp)
+ 176e: 00
+ 176f: 74 16 je 1787 <phase_3+0x14f>
+ 1771: e8 d4 05 00 00 call 1d4a <explode_bomb>
+ 1776: b8 79 00 00 00 mov $0x79,%eax
+ 177b: eb 0a jmp 1787 <phase_3+0x14f>
+ 177d: e8 c8 05 00 00 call 1d4a <explode_bomb>
+ 1782: b8 68 00 00 00 mov $0x68,%eax
+ 1787: 38 44 24 07 cmp %al,0x7(%rsp)
+ 178b: 75 05 jne 1792 <phase_3+0x15a>
+ 178d: 48 83 c4 18 add $0x18,%rsp
+ 1791: c3 ret
+ 1792: e8 b3 05 00 00 call 1d4a <explode_bomb>
+ 1797: eb f4 jmp 178d <phase_3+0x155>
+```
+
+```shell
+...
+ 165b: e8 80 fc ff ff call 12e0 <__isoc99_sscanf@plt>
+...
+```
+
+We can see that `scanf` is being called which means we need to figure out what datatype(s) the program is expecting.
+
+Because I do not want to enter the solutions to phases 1 and 2 again and again, I am goig to pass a file which has these solutions.
+
+```shell
+joxxxn@jupyter-nxxh6xx8:~/lab2-bomblab-navanchauhan/bombbomb$ gdb -ex 'break phase_3' -ex 'break explode_bomb' -ex 'run' -args ./bomb sol.txt
+GNU gdb (Ubuntu 12.1-0ubuntu1~22.04) 12.1
+Copyright (C) 2022 Free Software Foundation, Inc.
+License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html>
+This is free software: you are free to change and redistribute it.
+There is NO WARRANTY, to the extent permitted by law.
+Type "show copying" and "show warranty" for details.
+This GDB was configured as "x86_64-linux-gnu".
+Type "show configuration" for configuration details.
+For bug reporting instructions, please see:
+<https://www.gnu.org/software/gdb/bugs/>.
+Find the GDB manual and other documentation resources online at:
+ <http://www.gnu.org/software/gdb/documentation/>.
+
+For help, type "help".
+Type "apropos word" to search for commands related to "word"...
+Reading symbols from ./bomb...
+Breakpoint 1 at 0x1638
+Breakpoint 2 at 0x1d4a
+Starting program: /home/joxxxn/lab2-bomblab-navanchauhan/bombbomb/bomb sol.txt
+[Thread debugging using libthread_db enabled]
+Using host libthread_db library "/lib/x86_64-linux-gnu/libthread_db.so.1".
+Welcome to my fiendish little bomb. You have 6 phases with
+which to blow yourself up. Have a nice day!
+Phase 1 defused. How about the next one?
+That's number 2. Keep going!
+random string
+
+Breakpoint 1, 0x0000555555555638 in phase_3 ()
+(gdb) disas
+Dump of assembler code for function phase_3:
+=> 0x0000555555555638 <+0>: endbr64
+ 0x000055555555563c <+4>: sub $0x18,%rsp
+ 0x0000555555555640 <+8>: lea 0x7(%rsp),%rcx
+ 0x0000555555555645 <+13>: lea 0xc(%rsp),%rdx
+ 0x000055555555564a <+18>: lea 0x8(%rsp),%r8
+ 0x000055555555564f <+23>: lea 0x1b60(%rip),%rsi # 0x5555555571b6
+ 0x0000555555555656 <+30>: mov $0x0,%eax
+ 0x000055555555565b <+35>: call 0x5555555552e0 <__isoc99_sscanf@plt>
+ 0x0000555555555660 <+40>: cmp $0x2,%eax
+ 0x0000555555555663 <+43>: jle 0x555555555685 <phase_3+77>
+ 0x0000555555555665 <+45>: cmpl $0x7,0xc(%rsp)
+ 0x000055555555566a <+50>: ja 0x55555555577d <phase_3+325>
+ 0x0000555555555670 <+56>: mov 0xc(%rsp),%eax
+ 0x0000555555555674 <+60>: lea 0x1b55(%rip),%rdx # 0x5555555571d0
+ 0x000055555555567b <+67>: movslq (%rdx,%rax,4),%rax
+ 0x000055555555567f <+71>: add %rdx,%rax
+ 0x0000555555555682 <+74>: notrack jmp *%rax
+ 0x0000555555555685 <+77>: call 0x555555555d4a <explode_bomb>
+ 0x000055555555568a <+82>: jmp 0x555555555665 <phase_3+45>
+ 0x000055555555568c <+84>: mov $0x63,%eax
+ 0x0000555555555691 <+89>: cmpl $0x23d,0x8(%rsp)
+ 0x0000555555555699 <+97>: je 0x555555555787 <phase_3+335>
+ 0x000055555555569f <+103>: call 0x555555555d4a <explode_bomb>
+ 0x00005555555556a4 <+108>: mov $0x63,%eax
+ 0x00005555555556a9 <+113>: jmp 0x555555555787 <phase_3+335>
+--Type <RET> for more, q to quit, c to continue without paging--
+```
+
+
+`gdb` has thankfully marked the address which is being passed to `scanf`. We can access the value:
+
+```shell
+(gdb) x/1s 0x5555555571b6
+0x5555555571b6: "%d %c %d"
+(gdb)
+```
+
+BINGO! The program expects an integer, character, and another integer. Onwards.
+
+```shell
+ 0x0000555555555660 <+40>: cmp $0x2,%eax
+ 0x0000555555555663 <+43>: jle 0x555555555685 <phase_3+77>
+...
+ 0x0000555555555685 <+77>: call 0x555555555d4a <explode_bomb>
+```
+
+The program checks whether `scanf` returns a value <= 2, if it does then it calls the `explode_bomb` function.
+
+*Note: `scanf` returns the number of fields that were succesfully converted and assigned*
+
+```shell
+ 0x0000555555555665 <+45>: cmpl $0x7,0xc(%rsp)
+ 0x000055555555566a <+50>: ja 0x55555555577d <phase_3+325>
+...
+ 0x000055555555577d <+325>: call 0x555555555d4a <explode_bomb>
+```
+
+Similarly, the program checks and ensures the returned value is not > 7.
+
+
+```shell
+ 0x0000555555555670 <+56>: mov 0xc(%rsp),%eax
+ 0x0000555555555674 <+60>: lea 0x1b55(%rip),%rdx # 0x5555555571d0
+ 0x000055555555567b <+67>: movslq (%rdx,%rax,4),%rax
+ 0x000055555555567f <+71>: add %rdx,%rax
+ 0x0000555555555682 <+74>: notrack jmp *%rax
+ 0x0000555555555685 <+77>: call 0x555555555d4a <explode_bomb>
+```
+
+* `0x0000555555555670 <+56>: mov 0xc(%rsp),%eax` - Moves value located at `0xc` (12 in Decimal) bytes above the stack pointer to `%eax` register.
+* `0x0000555555555674 <+60>: lea 0x1b55(%rip),%rdx # 0x5555555571d0` - This instruction calculates an effective address by adding `0x1b55` to the current instruction pointer (`%rip`). The result is stored in the `%rdx` register.
+* `0x000055555555567b <+67>: movslq (%rdx,%rax,4),%rax`
+ * `movslq` stands for "move with sign-extension from a 32-bit value to a 64-bit value." (if the 32-bit value is negative, the 64-bit result will have all its upper 32 bits set to 1; otherwise, they'll be set to 0).
+ * `(%rdx,%rax,4)` - First start with the value in the %rdx register, then add to it the value in the %rax register multiplied by 4.
+ * `%rax` - Destination Register
+* `0x000055555555567f <+71>: add %rdx,%rax` - Adds base address in `%rdx` to the offset in `%rax`
+* `0x0000555555555682 <+74>: notrack jmp *%rax` - Jumps to the address stored in `%rax`
+* `0x0000555555555685 <+77>: call 0x555555555d4a <explode_bomb>` - If we are unable to jump to the specified instruction, call `explode_bomb`
+
+Let us try to run the program again with a valid input for the first number and see what the program is computing for the address.
+
+I used the input: `3 c 123`.
+
+To check what is the computed address, we can switch to the asm layout by running `layout asm`, and then going through instructions `ni` or `si` until we reach the line `movslq (%rdx,%rax,4),%rax`
+
+`%rax` should hold the value 3.
+
+```
+(gdb) print $rax
+$1 = 3
+```
+
+![Screenshot of GDB terminal depicting us checking the value of the instruction to be jumped to](/assets/bomb-lab/phase-3.png)
+
+We can see that this makes us jump to `<phase_3+186>` (Continue to step through the code by using `ni`)
+
+```shell
+ 0x00005555555556f2 <+186>: mov $0x64,%eax
+ 0x00005555555556f7 <+191>: cmpl $0x280,0x8(%rsp)
+ 0x00005555555556ff <+199>: je 0x555555555787 <phase_3+335>
+ 0x0000555555555705 <+205>: call 0x555555555d4a <explode_bomb>
+```
+
+We see that `0x64` (Decimal 100) is being stored in `%eax`. Then, the program compares `0x280` (Decimal 640) with memory address `0x8` bytes above the stack pointer (`%rsp`). If the values are equal, then it jumps to `<phase_3+335>`, otherwise `explode_bomb` is called.
+
+```shell
+ 0x0000555555555787 <+335>: cmp %al,0x7(%rsp)
+ 0x000055555555578b <+339>: jne 0x555555555792 <phase_3+346>
+ 0x000055555555578d <+341>: add $0x18,%rsp
+ 0x0000555555555791 <+345>: ret
+ 0x0000555555555792 <+346>: call 0x555555555d4a <explode_bomb>
+```
+
+Here, the program is comparing the value of our given character to the value stored in `%al` (lower 8 bits of `EAX`), and checks if they are not equal.
+
+Knowing that the character is stored at an offset of 7 bytes to `%rsp`, we can print and check the value by running:
+
+```shell
+(gdb) x/1cw $rsp+7
+c
+(gdb) print $al
+$1 = 100
+```
+
+We can simply lookup the [ASCII table](https://www.cs.cmu.edu/~pattis/15-1XX/common/handouts/ascii.html), and see that 100 in decimal stands for the character `d`. Let us try this answer:
+
+```shell
+...
+That's number 2. Keep going!
+3 d 640
+
+Breakpoint 1, 0x0000555555555638 in phase_3 ()
+(gdb) continue
+Continuing.
+Halfway there!
+```
+
+## Phase 4
+
+```shell
+joxxxn@jupyter-nxxh6xx8:~/lab2-bomblab-navanchauhan/bombbomb$ gdb -ex 'break phase_4' -ex 'break explode_bomb' -ex 'run' -args ./bomb sol.txt
+GNU gdb (Ubuntu 12.1-0ubuntu1~22.04) 12.1
+Copyright (C) 2022 Free Software Foundation, Inc.
+License GPLv3+: GNU GPL version 3 or later <http://gnu.org/licenses/gpl.html>
+This is free software: you are free to change and redistribute it.
+There is NO WARRANTY, to the extent permitted by law.
+Type "show copying" and "show warranty" for details.
+This GDB was configured as "x86_64-linux-gnu".
+Type "show configuration" for configuration details.
+For bug reporting instructions, please see:
+<https://www.gnu.org/software/gdb/bugs/>.
+Find the GDB manual and other documentation resources online at:
+ <http://www.gnu.org/software/gdb/documentation/>.
+
+For help, type "help".
+Type "apropos word" to search for commands related to "word"...
+Reading symbols from ./bomb...
+Breakpoint 1 at 0x17d3
+Breakpoint 2 at 0x1d4a
+Starting program: /home/joxxxn/lab2-bomblab-navanchauhan/bombbomb/bomb sol.txt
+[Thread debugging using libthread_db enabled]
+Using host libthread_db library "/lib/x86_64-linux-gnu/libthread_db.so.1".
+Welcome to my fiendish little bomb. You have 6 phases with
+which to blow yourself up. Have a nice day!
+Phase 1 defused. How about the next one?
+That's number 2. Keep going!
+Halfway there!
+test string
+
+Breakpoint 1, 0x00005555555557d3 in phase_4 ()
+(gdb) disas phase_4
+Dump of assembler code for function phase_4:
+=> 0x00005555555557d3 <+0>: endbr64
+ 0x00005555555557d7 <+4>: sub $0x18,%rsp
+ 0x00005555555557db <+8>: lea 0x8(%rsp),%rcx
+ 0x00005555555557e0 <+13>: lea 0xc(%rsp),%rdx
+ 0x00005555555557e5 <+18>: lea 0x1bba(%rip),%rsi # 0x5555555573a6
+ 0x00005555555557ec <+25>: mov $0x0,%eax
+ 0x00005555555557f1 <+30>: call 0x5555555552e0 <__isoc99_sscanf@plt>
+ 0x00005555555557f6 <+35>: cmp $0x2,%eax
+ 0x00005555555557f9 <+38>: jne 0x555555555802 <phase_4+47>
+ 0x00005555555557fb <+40>: cmpl $0xe,0xc(%rsp)
+ 0x0000555555555800 <+45>: jbe 0x555555555807 <phase_4+52>
+ 0x0000555555555802 <+47>: call 0x555555555d4a <explode_bomb>
+ 0x0000555555555807 <+52>: mov $0xe,%edx
+ 0x000055555555580c <+57>: mov $0x0,%esi
+ 0x0000555555555811 <+62>: mov 0xc(%rsp),%edi
+ 0x0000555555555815 <+66>: call 0x555555555799 <func4>
+ 0x000055555555581a <+71>: cmp $0x2,%eax
+ 0x000055555555581d <+74>: jne 0x555555555826 <phase_4+83>
+ 0x000055555555581f <+76>: cmpl $0x2,0x8(%rsp)
+ 0x0000555555555824 <+81>: je 0x55555555582b <phase_4+88>
+ 0x0000555555555826 <+83>: call 0x555555555d4a <explode_bomb>
+ 0x000055555555582b <+88>: add $0x18,%rsp
+ 0x000055555555582f <+92>: ret
+End of assembler dump.
+(gdb)
+```
+
+Again, `gdb` has marked the string being passed to `scanf`
+
+```shell
+(gdb) x/1s 0x5555555573a6
+0x5555555573a6: "%d %d"
+```
+
+Okay, so this time we are supposed to enter 2 numbers.
+
+```shell
+ 0x00005555555557f6 <+35>: cmp $0x2,%eax
+ 0x00005555555557f9 <+38>: jne 0x555555555802 <phase_4+47>
+```
+
+Checks if there were 2 values read from calling `scanf`, if not -> jump to `<phase_4+47>` which calls `<explode_bomb>`.
+
+```shell
+ 0x00005555555557fb <+40>: cmpl $0xe,0xc(%rsp)
+ 0x0000555555555800 <+45>: jbe 0x555555555807 <phase_4+52>
+```
+
+Compare `0xe` (14 in Decimal) and value stored at `$rsp` + `0xc` bytes (Decimal 12). If this condition is met (<= 14), jump to `<phase_4+52>`. If not, then explode bomb.
+
+```shell
+...
+ 0x0000555555555807 <+52>: mov $0xe,%edx
+ 0x000055555555580c <+57>: mov $0x0,%esi
+ 0x0000555555555811 <+62>: mov 0xc(%rsp),%edi
+ 0x0000555555555815 <+66>: call 0x555555555799 <func4>
+ 0x000055555555581a <+71>: cmp $0x2,%eax
+ 0x000055555555581d <+74>: jne 0x555555555826 <phase_4+83>
+ 0x000055555555581f <+76>: cmpl $0x2,0x8(%rsp)
+ 0x0000555555555824 <+81>: je 0x55555555582b <phase_4+88>
+ 0x0000555555555826 <+83>: call 0x555555555d4a <explode_bomb>
+```
+
+* ` 0x0000555555555815 <+66>: call 0x555555555799 <func4>` calls another function called `func4`
+* The returned value is compared with `0x2`, if they are not equal then the program jumps to call `<explode_bomb>`. This tells us that `func4` should return 2.
+
+Let us look into `func4`
+
+```shell
+(gdb) disas func4
+Dump of assembler code for function func4:
+ 0x0000555555555799 <+0>: endbr64
+ 0x000055555555579d <+4>: sub $0x8,%rsp
+ 0x00005555555557a1 <+8>: mov %edx,%ecx
+ 0x00005555555557a3 <+10>: sub %esi,%ecx
+ 0x00005555555557a5 <+12>: shr %ecx
+ 0x00005555555557a7 <+14>: add %esi,%ecx
+ 0x00005555555557a9 <+16>: cmp %edi,%ecx
+ 0x00005555555557ab <+18>: ja 0x5555555557b9 <func4+32>
+ 0x00005555555557ad <+20>: mov $0x0,%eax
+ 0x00005555555557b2 <+25>: jb 0x5555555557c5 <func4+44>
+ 0x00005555555557b4 <+27>: add $0x8,%rsp
+ 0x00005555555557b8 <+31>: ret
+ 0x00005555555557b9 <+32>: lea -0x1(%rcx),%edx
+ 0x00005555555557bc <+35>: call 0x555555555799 <func4>
+ 0x00005555555557c1 <+40>: add %eax,%eax
+ 0x00005555555557c3 <+42>: jmp 0x5555555557b4 <func4+27>
+ 0x00005555555557c5 <+44>: lea 0x1(%rcx),%esi
+ 0x00005555555557c8 <+47>: call 0x555555555799 <func4>
+ 0x00005555555557cd <+52>: lea 0x1(%rax,%rax,1),%eax
+ 0x00005555555557d1 <+56>: jmp 0x5555555557b4 <func4+27>
+```
+
+This looks like a recursive function :( (I hate recursive functions)
+
+Let's annotate the instructions.
+
+```shell
+endbr64
+sub $0x8,%rsp // subtract 8 bytes from the stack pointer
+mov %edx,%ecx // Move the value in register %edx to %ecx
+sub %esi,%ecx // Subtract the value in %esi from %ecx
+shr %ecx // Right shift the value in %ecx by one bit (dividing the value by 2)
+add %esi,%ecx // Add the value in %esi to %ecx
+cmp %edi,%ecx // Compare
+ja 0x5555555557b9 <func4+32> // If %ecx > %edi -> jump to instruction at offset +32
+mov $0x0,%eax // Move 0 to %eax
+jb 0x5555555557c5 <func4+44> // If %ecx < %edi -> jump to instruction at offset +44.
+add $0x8,%rsp // add 8 bytes to the stack pointer
+ret // return
+lea -0x1(%rcx),%edx // LEA of $rxc - 1 into $edx
+call 0x555555555799 <func4> // Call itself
+add %eax,%eax // Double the value in %eax
+jmp 0x5555555557b4 <func4+27> // jump to the instruction at offset +27
+lea 0x1(%rcx),%esi
+call 0x555555555799 <func4>
+lea 0x1(%rax,%rax,1),%eax // LEA of %rax * 2 + 1 into $eax
+jmp 0x5555555557b4 <func4+27>
+```
+
+We can either try to compute the values by hand, or write a simple script in Python to get the answer.
+
+```python
+def func4(edi, esi=0, edx=20):
+ ecx = (edx - esi) // 2 + esi
+ if ecx > edi:
+ return 2 * func4(edi, esi, ecx - 1)
+ elif ecx < edi:
+ return 2 * func4(edi, ecx + 1, edx) + 1
+ else:
+ return 0
+
+for x in range(15): # We can limit to 14
+ if func4(x) == 2:
+ print(f"answer is {x}")
+ break
+```
+
+Running this code, we get: `answer is 5`
+
+Okay, so we know that the number needed to be passed to `func4` is 5. But, what about the second digit?
+
+If we go back to the code for `<phase_4>`, we can see that:
+
+```shell
+ 0x000055555555581f <+76>: cmpl $0x2,0x8(%rsp)
+ 0x0000555555555824 <+81>: je 0x55555555582b <phase_4+88>
+```
+
+The value at `$rsp+8` should be equal to 2. So, let us try passing `5 2` as our input.
+
+```shell
+...
+Phase 1 defused. How about the next one?
+That's number 2. Keep going!
+Halfway there!
+5 2
+
+Breakpoint 1, 0x00005555555557d3 in phase_4 ()
+(gdb) continue
+Continuing.
+So you got that one. Try this one.
+```
+
+## Phase 5
+
+```shell
+So you got that one. Try this one.
+test string
+
+Breakpoint 1, 0x0000555555555830 in phase_5 ()
+(gdb) disas phase_5
+Dump of assembler code for function phase_5:
+=> 0x0000555555555830 <+0>: endbr64
+ 0x0000555555555834 <+4>: push %rbx
+ 0x0000555555555835 <+5>: sub $0x10,%rsp
+ 0x0000555555555839 <+9>: mov %rdi,%rbx
+ 0x000055555555583c <+12>: call 0x555555555b10 <string_length>
+ 0x0000555555555841 <+17>: cmp $0x6,%eax
+ 0x0000555555555844 <+20>: jne 0x55555555588b <phase_5+91>
+ 0x0000555555555846 <+22>: mov $0x0,%eax
+ 0x000055555555584b <+27>: lea 0x199e(%rip),%rcx # 0x5555555571f0 <array.0>
+ 0x0000555555555852 <+34>: movzbl (%rbx,%rax,1),%edx
+ 0x0000555555555856 <+38>: and $0xf,%edx
+ 0x0000555555555859 <+41>: movzbl (%rcx,%rdx,1),%edx
+ 0x000055555555585d <+45>: mov %dl,0x9(%rsp,%rax,1)
+ 0x0000555555555861 <+49>: add $0x1,%rax
+ 0x0000555555555865 <+53>: cmp $0x6,%rax
+ 0x0000555555555869 <+57>: jne 0x555555555852 <phase_5+34>
+ 0x000055555555586b <+59>: movb $0x0,0xf(%rsp)
+ 0x0000555555555870 <+64>: lea 0x9(%rsp),%rdi
+ 0x0000555555555875 <+69>: lea 0x1943(%rip),%rsi # 0x5555555571bf
+ 0x000055555555587c <+76>: call 0x555555555b31 <strings_not_equal>
+ 0x0000555555555881 <+81>: test %eax,%eax
+ 0x0000555555555883 <+83>: jne 0x555555555892 <phase_5+98>
+ 0x0000555555555885 <+85>: add $0x10,%rsp
+ 0x0000555555555889 <+89>: pop %rbx
+ 0x000055555555588a <+90>: ret
+ 0x000055555555588b <+91>: call 0x555555555d4a <explode_bomb>
+ 0x0000555555555890 <+96>: jmp 0x555555555846 <phase_5+22>
+ 0x0000555555555892 <+98>: call 0x555555555d4a <explode_bomb>
+ 0x0000555555555897 <+103>: jmp 0x555555555885 <phase_5+85>
+End of assembler dump.
+(gdb)
+```
+
+```shell
+...
+ 0x000055555555583c <+12>: call 0x555555555b10 <string_length>
+ 0x0000555555555841 <+17>: cmp $0x6,%eax
+ 0x0000555555555844 <+20>: jne 0x55555555588b <phase_5+91>
+...
+ 0x000055555555588b <+91>: call 0x555555555d4a <explode_bomb>
+...
+```
+
+First things first, these instructions check to make sure the passed string is of length 6, otherwise `explode_bomb` is called.
+
+We can also see a similar pattern compared to Phase 2, where we had a loop:
+
+* The looping part:
+ * `mov $0x0,%eax` - Initialise `%eax` and set it to 0 (our counter/iterator)
+ * `movzbl (%rbx,%rax,1),%edx` - Access `%rbx + 1 * %rax` and store it in `%edx`
+ * `and $0xf,%edx` - Take the least significant 4 bits of the byte.
+ * `movzbl (%rcx,%rdx,1),%edx` - Use the 4 bits as an index into another array and load the corresponding byte into `%edx`
+ * `mov %dl,0x9(%rsp,%rax,1)` - Store the transformed byte into a buffer on the stack
+ * `add $0x1,%rax` - Increment `%rax`
+ * `cmp $0x6,%rax` - If the index is not yet 6, loop again
+* `movb $0x0,0xf(%rsp)` - Null-terminate the transformed string
+* `lea 0x9(%rsp),%rdi` and `lea 0x1943(%rip),%rsi`
+* `all 0x555555555b31 <strings_not_equal>` check if the two strings loaded up just before this are equal or not.
+
+We can check the reference string we need, which `gdb` has marked as `# 0x5555555571bf`, and the lookup table marked as `# 0x5555555571f0 <array.0>`
+
+```shell
+(gdb) x/s 0x5555555571bf
+0x5555555571bf: "bruins"
+(gdb) x/s 0x5555555571f0
+0x5555555571f0 <array.0>: "maduiersnfotvbylSo you think you can stop the bomb with ctrl-c, do you?"
+(gdb)
+```
+
+To summarize the transformation process:
+
+* The function takes each byte of the string
+* It keeps only the least significant 4 bits of each byte
+* It uses these 4 bits as an index into the lookup table (`array.0`)
+* The value from the array is then stored in a buffer
+
+Here's how the transformation process can be reversed for each character in "bruins":
+1. Find the index of `b` in the lookup table (in our case, it is 13 since we index starting 0)
+2. Calculate binary representation of this index (in our case 13 can be written as 1101 in binary)
+3. Find ASCII character whose least significant 4 bits match (in our case, `m` has binary representation `01101101`)
+
+Repeat for all 6 characters
+
+*Hint: Using an [ASCII - Binary Table](http://sticksandstones.kstrom.com/appen.html) can save you time.*
+
+Thus, we can have the following transformation:
+
+```
+b -> m
+r -> f
+u -> c
+i -> d
+n -> h
+s -> g
+```
+
+
+Let us try out this answer:
+
+```shell
+...
+That's number 2. Keep going!
+Halfway there!
+So you got that one. Try this one.
+mfcdhg
+
+Breakpoint 1, 0x0000555555555830 in phase_5 ()
+(gdb) continue
+Continuing.
+Good work! On to the next...
+```
+
+Awesome!
+
+## Phase 6
+
+```shell
+Good work! On to the next...
+test string
+
+Breakpoint 1, 0x0000555555555899 in phase_6 ()
+(gdb) disas phase_6
+Dump of assembler code for function phase_6:
+=> 0x0000555555555899 <+0>: endbr64
+ 0x000055555555589d <+4>: push %r15
+ 0x000055555555589f <+6>: push %r14
+ 0x00005555555558a1 <+8>: push %r13
+ 0x00005555555558a3 <+10>: push %r12
+ 0x00005555555558a5 <+12>: push %rbp
+ 0x00005555555558a6 <+13>: push %rbx
+ 0x00005555555558a7 <+14>: sub $0x68,%rsp
+ 0x00005555555558ab <+18>: lea 0x40(%rsp),%rax
+ 0x00005555555558b0 <+23>: mov %rax,%r14
+ 0x00005555555558b3 <+26>: mov %rax,0x8(%rsp)
+ 0x00005555555558b8 <+31>: mov %rax,%rsi
+ 0x00005555555558bb <+34>: call 0x555555555d97 <read_six_numbers>
+ 0x00005555555558c0 <+39>: mov %r14,%r12
+ 0x00005555555558c3 <+42>: mov $0x1,%r15d
+ 0x00005555555558c9 <+48>: mov %r14,%r13
+ 0x00005555555558cc <+51>: jmp 0x555555555997 <phase_6+254>
+ 0x00005555555558d1 <+56>: call 0x555555555d4a <explode_bomb>
+ 0x00005555555558d6 <+61>: jmp 0x5555555559a9 <phase_6+272>
+ 0x00005555555558db <+66>: add $0x1,%rbx
+ 0x00005555555558df <+70>: cmp $0x5,%ebx
+ 0x00005555555558e2 <+73>: jg 0x55555555598f <phase_6+246>
+ 0x00005555555558e8 <+79>: mov 0x0(%r13,%rbx,4),%eax
+ 0x00005555555558ed <+84>: cmp %eax,0x0(%rbp)
+ 0x00005555555558f0 <+87>: jne 0x5555555558db <phase_6+66>
+ 0x00005555555558f2 <+89>: call 0x555555555d4a <explode_bomb>
+ 0x00005555555558f7 <+94>: jmp 0x5555555558db <phase_6+66>
+ 0x00005555555558f9 <+96>: mov 0x8(%rsp),%rdx
+ 0x00005555555558fe <+101>: add $0x18,%rdx
+ 0x0000555555555902 <+105>: mov $0x7,%ecx
+ 0x0000555555555907 <+110>: mov %ecx,%eax
+ 0x0000555555555909 <+112>: sub (%r12),%eax
+ 0x000055555555590d <+116>: mov %eax,(%r12)
+ 0x0000555555555911 <+120>: add $0x4,%r12
+ 0x0000555555555915 <+124>: cmp %r12,%rdx
+ 0x0000555555555918 <+127>: jne 0x555555555907 <phase_6+110>
+ 0x000055555555591a <+129>: mov $0x0,%esi
+ 0x000055555555591f <+134>: mov 0x40(%rsp,%rsi,4),%ecx
+ 0x0000555555555923 <+138>: mov $0x1,%eax
+ 0x0000555555555928 <+143>: lea 0x3d01(%rip),%rdx # 0x555555559630 <node1>
+--Type <RET> for more, q to quit, c to continue without paging--
+ 0x000055555555592f <+150>: cmp $0x1,%ecx
+ 0x0000555555555932 <+153>: jle 0x55555555593f <phase_6+166>
+ 0x0000555555555934 <+155>: mov 0x8(%rdx),%rdx
+ 0x0000555555555938 <+159>: add $0x1,%eax
+ 0x000055555555593b <+162>: cmp %ecx,%eax
+ 0x000055555555593d <+164>: jne 0x555555555934 <phase_6+155>
+ 0x000055555555593f <+166>: mov %rdx,0x10(%rsp,%rsi,8)
+ 0x0000555555555944 <+171>: add $0x1,%rsi
+ 0x0000555555555948 <+175>: cmp $0x6,%rsi
+ 0x000055555555594c <+179>: jne 0x55555555591f <phase_6+134>
+ 0x000055555555594e <+181>: mov 0x10(%rsp),%rbx
+ 0x0000555555555953 <+186>: mov 0x18(%rsp),%rax
+ 0x0000555555555958 <+191>: mov %rax,0x8(%rbx)
+ 0x000055555555595c <+195>: mov 0x20(%rsp),%rdx
+ 0x0000555555555961 <+200>: mov %rdx,0x8(%rax)
+ 0x0000555555555965 <+204>: mov 0x28(%rsp),%rax
+ 0x000055555555596a <+209>: mov %rax,0x8(%rdx)
+ 0x000055555555596e <+213>: mov 0x30(%rsp),%rdx
+ 0x0000555555555973 <+218>: mov %rdx,0x8(%rax)
+ 0x0000555555555977 <+222>: mov 0x38(%rsp),%rax
+ 0x000055555555597c <+227>: mov %rax,0x8(%rdx)
+ 0x0000555555555980 <+231>: movq $0x0,0x8(%rax)
+ 0x0000555555555988 <+239>: mov $0x5,%ebp
+ 0x000055555555598d <+244>: jmp 0x5555555559c4 <phase_6+299>
+ 0x000055555555598f <+246>: add $0x1,%r15
+ 0x0000555555555993 <+250>: add $0x4,%r14
+ 0x0000555555555997 <+254>: mov %r14,%rbp
+ 0x000055555555599a <+257>: mov (%r14),%eax
+ 0x000055555555599d <+260>: sub $0x1,%eax
+ 0x00005555555559a0 <+263>: cmp $0x5,%eax
+ 0x00005555555559a3 <+266>: ja 0x5555555558d1 <phase_6+56>
+ 0x00005555555559a9 <+272>: cmp $0x5,%r15d
+ 0x00005555555559ad <+276>: jg 0x5555555558f9 <phase_6+96>
+ 0x00005555555559b3 <+282>: mov %r15,%rbx
+ 0x00005555555559b6 <+285>: jmp 0x5555555558e8 <phase_6+79>
+ 0x00005555555559bb <+290>: mov 0x8(%rbx),%rbx
+ 0x00005555555559bf <+294>: sub $0x1,%ebp
+ 0x00005555555559c2 <+297>: je 0x5555555559d5 <phase_6+316>
+ 0x00005555555559c4 <+299>: mov 0x8(%rbx),%rax
+ 0x00005555555559c8 <+303>: mov (%rax),%eax
+ 0x00005555555559ca <+305>: cmp %eax,(%rbx)
+--Type <RET> for more, q to quit, c to continue without paging--
+ 0x00005555555559cc <+307>: jge 0x5555555559bb <phase_6+290>
+ 0x00005555555559ce <+309>: call 0x555555555d4a <explode_bomb>
+ 0x00005555555559d3 <+314>: jmp 0x5555555559bb <phase_6+290>
+ 0x00005555555559d5 <+316>: add $0x68,%rsp
+ 0x00005555555559d9 <+320>: pop %rbx
+ 0x00005555555559da <+321>: pop %rbp
+ 0x00005555555559db <+322>: pop %r12
+ 0x00005555555559dd <+324>: pop %r13
+ 0x00005555555559df <+326>: pop %r14
+ 0x00005555555559e1 <+328>: pop %r15
+ 0x00005555555559e3 <+330>: ret
+End of assembler dump.
+(gdb)
+```
+
+Again, we see the familiar `read_six_digits` function.
+
+Let us analyse this function in chunks:
+
+```shell
+ 0x00005555555558bb <+34>: call 0x555555555d97 <read_six_numbers>
+ 0x00005555555558c0 <+39>: mov %r14,%r12
+ 0x00005555555558c3 <+42>: mov $0x1,%r15d
+ 0x00005555555558c9 <+48>: mov %r14,%r13
+ 0x00005555555558cc <+51>: jmp 0x555555555997 <phase_6+254>
+```
+
+1. Read six numbers
+2. Initialise Registers:
+ 2.1. `mov %r14,%r12`: `%r14` should be pointing to the location of the stack where the numbers were read into. This address is copied onto `%r12`
+ 2.2. `mov $0x1,%r15d`: The value `1` is moved into `%r15` register (probably acting like a counter)
+ 2.3. `mov %r14,%r13`: The value is also copied to `%r13`
+3. Jump to start of loop:
+
+```shell
+ 0x0000555555555997 <+254>: mov %r14,%rbp
+ 0x000055555555599a <+257>: mov (%r14),%eax
+ 0x000055555555599d <+260>: sub $0x1,%eax
+ 0x00005555555559a0 <+263>: cmp $0x5,%eax
+ 0x00005555555559a3 <+266>: ja 0x5555555558d1 <phase_6+56>
+```
+
+1. Initialise register and point to first number in sequence
+2. Adjust number(s):
+ 2.1. `mov (%r14),%eax` -> load the current number in the sequence
+ 2.2. `sub $0x1,%eax` -> decrement number by 1
+3. Validation
+ 3.1. `cmp $0x5,%eax`: This compares the adjusted value in `%eax` with 5.
+ 3.2. `ja 0x5555555558d1 <phase_6+56>`: jump if given value is > 5 or < 0
+
+=> All numbers should be between 1 and 6.
+
+```shell
+ 0x00005555555559a9 <+272>: cmp $0x5,%r15d
+ 0x00005555555559ad <+276>: jg 0x5555555558f9 <phase_6+96>
+```
+
+This checks if the value stored in `%r15` is > 5, if it is then it jumps somewhere else. This validates our assumption that `%r15` is acting as a counter.
+
+```shell
+ 0x00005555555559b3 <+282>: mov %r15,%rbx
+ 0x00005555555559b6 <+285>: jmp 0x5555555558e8 <phase_6+79>
+```
+
+Let us jump to +79
+
+```shell
+ 0x00005555555558e8 <+79>: mov 0x0(%r13,%rbx,4),%eax
+ 0x00005555555558ed <+84>: cmp %eax,0x0(%rbp)
+ 0x00005555555558f0 <+87>: jne 0x5555555558db <phase_6+66>
+ 0x00005555555558f2 <+89>: call 0x555555555d4a <explode_bomb>
+ 0x00005555555558f7 <+94>: jmp 0x5555555558db <phase_6+66>
+```
+
+This section deals with checking if all the numbers in the sequence are unique or not. Thus, we need to ensure out 6 digits are unique
+
+```shell
+ 0x00005555555558db <+66>: add $0x1,%rbx // Increments by 1
+ 0x00005555555558df <+70>: cmp $0x5,%ebx
+ 0x00005555555558e2 <+73>: jg 0x55555555598f <phase_6+246> // Jump if > 5 (Loop iterations are complete)
+ 0x00005555555558e8 <+79>: mov 0x0(%r13,%rbx,4),%eax
+ 0x00005555555558ed <+84>: cmp %eax,0x0(%rbp)
+ 0x00005555555558f0 <+87>: jne 0x5555555558db <phase_6+66> // Again, check if the number being seen is unique
+```
+
+Now we know that the numbers are unique, between 1-6 (inclusive).
+
+After stepping through the instructions, we can also see that the numbers are being transformed:
+* By subtracting it from 7 (mov $0x7,%ecx followed by sub (%r12),%eax)
+* This effectively maps the numbers as follows: 1 to 6, 2 to 5, 3 to 4, 4 to 3, 5 to 2, and 6 to 1.
+
+Let us try to figure out what ` 0x0000555555555928 <+143>: lea 0x3d01(%rip),%rdx # 0x555555559630 <node1>` is:
+
+```shell
+(gdb) x/30wx 0x555555559630
+0x555555559630 <node1>: 0x000000d9 0x00000001 0x55559640 0x00005555
+0x555555559640 <node2>: 0x000003ab 0x00000002 0x55559650 0x00005555
+0x555555559650 <node3>: 0x0000014f 0x00000003 0x55559660 0x00005555
+0x555555559660 <node4>: 0x000000a1 0x00000004 0x55559670 0x00005555
+0x555555559670 <node5>: 0x000001b3 0x00000005 0x55559120 0x00005555
+0x555555559680 <host_table>: 0x555573f5 0x00005555 0x5555740f 0x00005555
+0x555555559690 <host_table+16>: 0x55557429 0x00005555 0x00000000 0x00000000
+0x5555555596a0 <host_table+32>: 0x00000000 0x00000000
+(gdb) x/30wx 0x555555559120
+0x555555559120 <node6>: 0x000002da 0x00000006 0x00000000 0x00000000
+0x555555559130: 0x00000000 0x00000000 0x00000000 0x00000000
+0x555555559140 <userid>: 0x61767861 0x38383535 0x00000000 0x00000000
+0x555555559150 <userid+16>: 0x00000000 0x00000000 0x00000000 0x00000000
+0x555555559160 <userid+32>: 0x00000000 0x00000000 0x00000000 0x00000000
+0x555555559170 <userid+48>: 0x00000000 0x00000000 0x00000000 0x00000000
+0x555555559180 <userid+64>: 0x00000000 0x00000000 0x00000000 0x00000000
+0x555555559190 <userid+80>: 0x00000000 0x00000000
+(gdb)
+```
+
+It appears that this is a linked list. With roughly the following structure:
+
+```cpp
+struct node {
+ int value;
+ int index;
+ struct node *next;
+};
+```
+
+Let us convert the values into decimal:
+
+```
+0x000000d9 -> 217
+0x000003ab -> 939
+0x0000014f -> 335
+0x000000a1 -> 161
+0x000001b3 -> 435
+0x000002da -> 730
+```
+
+**Missing Notes**
+
+To re-arrange this linked list in descending order, we would arrange it as follows:
+
+```
+Node 2 -> Node 6 -> Node 5 -> Node 3 -> Node 1 -> Node 4
+```
+
+Since we also need to apply the transformation: `7 - x`:
+
+```
+(7-2) -> (7-6) -> ... -> (7-4)
+```
+
+Final answer: `5 1 2 4 6 3`
+
+Let us try the answer:
+
+```
+...
+That's number 2. Keep going!
+Halfway there!
+So you got that one. Try this one.
+Good work! On to the next...
+5 1 2 4 6 3
+
+Breakpoint 1, 0x0000555555555899 in phase_6 ()
+(gdb) continue
+Continuing.
+Congratulations! You've defused the bomb!
+Your instructor has been notified and will verify your solution.
+[Inferior 1 (process 1754) exited normally]
+```
+
+But, what about the secret phase? \ No newline at end of file
diff --git a/Content/posts/2023-10-05-attack-lab.md b/Content/posts/2023-10-05-attack-lab.md
new file mode 100644
index 0000000..cfd7bd8
--- /dev/null
+++ b/Content/posts/2023-10-05-attack-lab.md
@@ -0,0 +1,22 @@
+---
+date: 2023-10-05 20:01
+description: Walkthrough of Attack Lab for CSCI 2400 Computer Systems
+tags: gdb, reverse-engineering, c++, csci2400, assembly
+draft: true
+---
+
+# Attack Lab
+
+## Introduction
+
+Lab 3 for CSCI 2400 @ CU Boulder - Computer Systems
+
+> This assignment involves generating a total of five attacks on two programs having different security vulnerabilities. The directions for this lab are detailed but not difficult to follow.
+
+Again, I like using objdump to disassemble the code.
+
+`objdump -d ctarget > dis.txt`
+
+## Phase 1
+
+From the instructions, we know that our task is to get `CTARGET` to execute the code for `touch1` when `getbuf` executes its return statement, rather than returning to `test`