text
stringlengths
0
1.99k
-------------------------------------
mach_port_t               connection:        0x1d07
uint32_t                  selector:          0x0
const uint64_t *          input:             0x0
uint32_t                  inputCnt:          0x0
const void *              inputStruct:       0x600001310240
size_t                    inputStructCnt:    0x60
uint64_t *                output:            0x0
uint32_t *                outputCnt:         0x16fdfe994
void *                    outputStruct:      0x16fdfe998
size_t *                  outputStructCnt:   0x100068a20
Input Struct Hexdump (first 32 bytes):
00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
00 00 00 00 00 00 00 00  00 00 00 00 00 00 00 00  |................|
````````````````````````````````````````````````````````````````````````````
However, the most handy function I made is iokit_dump. By using this, we
can gather corpus for fuzzer:
````````````````````````````````````````````````````````````````````````````
(lldb) iokit_dump corpus.bin
Successfully dumped 96 bytes of inputStruct to 'corpus.bin'
````````````````````````````````````````````````````````````````````````````
======
--[ 2.2.5 - Tracer
============================================================================
One more thing. We can also upgrade the previously introduced
trace_ioserviceopen.py to monitor the IOConnectCallMethod, so we have the
whole picture of what process opens a connection to which service using
what type of user client with what selector number, and finally, the proper
sizes of the arguments. So here it is, the iokit_tracer.py[38]:
````````````````````````````````````````````````````````````````````````````
command script import iokit_tracer.py
trace_iokit --pid 81203
----------------------------------------------------------------
PID: 81203
EXE PATH: /System/Library/CoreServices/Finder.app/Contents/MacOS/Finder
SERVICE: AppleAPFSContainer (Connection: 0x21267)
TYPE: 0
SELECTOR: 8 (0x8)
inputCnt: 0
inputStruct: 0x0
inputStructCnt: 0
output: 0x0
outputCnt: 0x0
outputStruct: 0x0
outputStructCnt: 0x0
----------------------------------------------------------------
trace_iokit --executable_path --executable_path crims0n -- -arg1 val1
----------------------------------------------------------------
PID: 8236 | EXE: /Users/karmaz/r/scripts/FUZZER/CRIMS0N/bin/crims0n
SERVICE: H11ANE (Connection: 0x1a07) | TYPE: 1
SELECTOR: 0 (0x0)
--- INPUT ---
input Scalars (2 values at 0x600000330020): [0x123, 0x123]
inputStruct (96 bytes at 0x600002534240):
0000: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 |................|
0010: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 |................|
0020: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 |................|
0030: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 |................|
0040: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 |................|
0050: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 |................|
--- OUTPUT ---
outputCnt: 0 (capacity pointer: 0x16b08e974)
outputStructCnt: 0 (capacity pointer: 0x104dd8a10)
````````````````````````````````````````````````````````````````````````````
This effectively finishes the runtime mapping part. How do we verify that
the data we gather through all of these guides is valid, though?
================
--[ 2.3 - Map Verification
============================================================================
Finally, a tool of the trade when it comes down to verification of our
map. IOVerify[39]. I created it a long time ago as part of my fuzzer. I
decided to extract the logic of it to make it a separate tool that can be
used for testing external methods user clients types, matching services,
etc. Bascially all of the things we learned here.
I recommend checking out the code. Here is the helper:
````````````````````````````````````````````````````````````````````````````
❯ ./IOVerify -h
Usage: ./IOVerify -n <name> (-m <method> | -y <spec>) [options]
Options:
-n <name> Target driver class name (required).
-t <type> Connection type (default: 0).
-m <id> Method selector ID.
-y <spec> Specify method and buffer sizes in one string.
Format: "ID: [IN_SCA, IN_STR, OUT_SCA, OUT_STR]"
Example: -y "0: [0, 96, 0, 96]"
-p <string> Payload as a string.
-f <file> File path for payload.
-b <hex_str> Space-separated hex string payload.
-i <size> Input buffer size (ignored if -y is used).
-o <size> Output buffer size (ignored if -y is used).
-s <value> Scalar input (uint64_t). Can be specified multiple times.
-S <count> Scalar output count (ignored if -y is used).
-h Show this help message.
````````````````````````````````````````````````````````````````````````````
And here is the most common usage so we can check if we enumerated properly