Jump to content

Simplified MSHookMemory Usage [ARM64 Only]


23 posts in this topic

Recommended Posts

Updated (edited)

While MSHookMemory is great, I hate I have to do this:

const uint8_t hack[] = {
    0x00, 0x00, 0x80, 0x52, // mov w0, #0
    0xc0, 0x03, 0x5f, 0xd6, // ret
};

 

 

So I wrote some code which won't need you to do that:

#include <substrate.h>
#include <mach-o/dyld.h>

#define ASLR_BIAS _dyld_get_image_vmaddr_slide(0)

uint64_t getRealOffset(uint64_t offset){
	return ASLR_BIAS + offset;
}

// main func
void inject(uint64_t offset, uint64_t hackedHex) {

    hackedHex = CFSwapInt32(hackedHex);        

    MSHookMemory((void *)getRealOffset(offset), (void *)&hackedHex, sizeof(hackedHex));
}

 

How to import this code to your project?

You have two options:

1. Copy & paste the code from above under your "imports"

2. Download this file & paste it in /var/theos/includes & write this on top of tweak.xm:

#include <inject.h>

 

Usage of the function:

//parameters it takes
inject(0xIDAOffset, 0xHackedHex);

//actual usage on a offset
inject(0x100299DC4, 0xC0035FD6);

 

 

Bugs to be fixed:

1. If you wanna write a hex more than 4 bytes, it won't work the way you think, example:

inject(0x100299DC0, 0x20008052C0035FD6);

/*

	This will write C0035FD620008052 instead of what you entered due the CFSwapInt32

/*

NOTE:  I'm no pro at this kinda stuff (dealing with memory), so if there's anything just mention it ^^

 

Credits

- Saurik

- topics from https://stackoverflow.com/   which helped me understand several things

- Me for this simplified version

Updated by Joey
  • Like 3
  • Winner 3
  • Informative 1
Posted

is it possible to make the hackedHex param a stirng, make it an array of strings which gets split at every 2nd character and add a 0x to each of them

Posted (edited)
  On 3/5/2019 at 1:23 AM, TheArmKing said:

is it possible to make the hackedHex param a stirng, make it an array of strings which gets split at every 2nd character and add a 0x to each of them

Expand  

Yeah I tried that, but couldn’t get something like it to work. I’ve tried multiple things, this is the only one that worked. Will look into an alternative another time. C/C++ isn’t the most user friendly when it comes to arrays it seems 

Updated by Joey
Posted (edited)
  On 3/5/2019 at 1:40 AM, Joey said:

Yeah I tried that, but couldn’t get something like it to work. I’ve tried multiple things, this is the only one that worked. Will look into an alternative another time. C/C++ isn’t the most user friendly when it comes to arrays it seems 

Expand  

something like this maybe?
 

// Convert long long to string
// Convert string to char array
// loop through char array to make strings of 2 characters or 1 byte, like c00fe3d4 becomes c0 0f e3 d4
// c0 gets converted to int
// the int gets converted to uint8
// the uint8 gets appeneded to aa uint8 array
// copied this from stackexchange 
char str[256];
sprintf(str, "%lld", hackedHex);
printf("%s\n", str); 

for (int x=0; x<str.length; x++ ) { //dk if this is how C works
  if(x%2==0){
    char a = str[x]
  }
  if(x%2==1){
    char b = str[x]
    // do something that combines char a and b to a string
    // then convert to int
    // then to uint
    // then append to array
}


 

Updated by TheArmKing
added some bs
Posted

I don't know why people prefer integers type for this purpose which forces you to deal with endianness.

I just simplified yours too.

void inject(uint64_t offset, const void *bytes, size_t bytes_len) {
	 MSHookMemory((void *)getRealOffset(offset), bytes, bytes_len);
}

Usage:

//offset & bytes literal & bytes length
/* 
mov w0, #1
ret 
*/
inject(0x100299DC4, "\x20\x00\x80\x52\xC0\x03\x5F\xD6", 8);

You can write as many bytes as you want with this.

Posted
  On 3/5/2019 at 3:25 AM, CyberCat said:

I don't know why people prefer integers type for this purpose which forces you to deal with endianness.

I just simplified yours too.

void inject(uint64_t offset, const void *bytes, size_t bytes_len) {
	 MSHookMemory((void *)getRealOffset(offset), bytes, bytes_len);
}

Usage:

//offset & bytes literal & bytes length
/* 
mov w0, #1
ret 
*/
inject(0x100299DC4, "\x20\x00\x80\x52\xC0\x03\x5F\xD6", 8);

You can write as many bytes as you want with this.

Expand  

The thing is, I hate to have to write it like that ^, it’s almost the same as the original now :p 

  On 3/5/2019 at 2:04 AM, TheArmKing said:

something like this maybe?
 

// Convert long long to string
// Convert string to char array
// loop through char array to make strings of 2 characters or 1 byte, like c00fe3d4 becomes c0 0f e3 d4
// c0 gets converted to int
// the int gets converted to uint8
// the uint8 gets appeneded to aa uint8 array
// copied this from stackexchange 
char str[256];
sprintf(str, "%lld", hackedHex);
printf("%s\n", str); 

for (int x=0; x<str.length; x++ ) { //dk if this is how C works
  if(x%2==0){
    char a = str[x]
  }
  if(x%2==1){
    char b = str[x]
    // do something that combines char a and b to a string
    // then convert to int
    // then to uint
    // then append to array
}


 

Expand  

possibly, but I think there must be a simpler solution 

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
  • Create New...

Important Information

We would like to place cookies on your device to help make this website better. The website cannot give you the best user experience without cookies. You can accept or decline our cookies. You may also adjust your cookie settings. Privacy Policy - Guidelines