qwertyoruiop

I have asked on Twitter if anyone could sign it for me. For some reason neither of the two people who tried to do so were able to sign it. No idea why. kexts were signed but they kept getting rejected for some reason.

iOS is vulnerable too as far as the vulnerability is concerned. It is not directly exploitable on iOS, however having a NULL task_t still does give you some abilities, even if not (directly?) SVC code exec.

while I agree with you on the security benefits of a full microkernel, to be entirely honest, if you had access to just IOKit you could easily use a network card or an hard drive controller to get a physical memory write-what-where, which in turn would allow you to gain access to anything, plus the microkernel performance issues of e.g. having to context switch on interrupts.

well, this bug is a null pointer deference. smap is like -no_shared_cr3, but without the performance loss.

Then you should not install them on a machine with confidential data at all, source or binary. It's that simple!

for the record: "At 0x20 I place a POP RAX;RET gadget" should be "At 0x18 I place a POP RAX;RET gadget".

for the record: i had no idea yesterday was saturday at the time I dropped the code.

You are not vulnerable since you have SMAP!

If the heap info leak fails, I bail out cleanly. If the kASLR leak fails, it is usually because instead of hitting a vm_map_copy (the intended structure I need to corrupt), something completely unrelated is hit instead. If it happens to hit something different than expected, that's undefined behaviour usually ending up in a panic.

1) I cannot really discuss specifics, but this particular bug would have been hard to find via a traditional IOKit fuzz, since it requires an invalid 'task' port passed over to IOServiceOpen. Most fuzzers use mach_task_self for that, and fuzz method calls/traps/properties/etc.

2) When IOServiceRelease is called, vtable+0x20 is called. the vtable pointer is controlled, at +0x20 I place a stack pivot, which sets RSP = RAX and pops 3 times. At 0x20 I place a POP RAX;RET gadget to let the chain begin after 0x28. Payload then locates the credentials structure, sets UID to 0 by bzero()ing, cleans up the memory corruption, decreases the task count for current user and increases task count for root. It then unlocks locks held by IOAudioEngine to prevent your audio from freezing up, and then returns to the userland context.

'sudo nvram boot-args=-no_shared_cr3' will do the trick.

The flag essentially prevents kernel from accessing userland memory unless special routines are used. Since the bug is a NULL pointer deference (which requires a read to userland memory in order to be exploited), exploitation becomes impossible. Due to this flag, however, your kernel will have to context switch every time a system call is done, which does have a noticeable performance impact. I will be releasing a KEXT to fix the bug soon.

Yes, it is.