Apple Exclaves • Hacker News Board

Re: Apple Exclaves

#231

Post by saagarjha » Sat, Mar 15, 2025, 10:36 AM UTC

hedora wrote: post #214

lxgr wrote: post #212

Earlier quoted context omitted.

How? The secure enclave doesn't have any trusted input/output capabilities other than the power button double click and biometrics. The PIN isn't entered via these.

Also, can't you just get the pin from the accelerometer? (It may take a few tries, but this was shown by academics a while back.)

My understanding is that sensors are disabled during this

Re: Apple Exclaves

#232

Post by saagarjha » Sat, Mar 15, 2025, 10:37 AM UTC

lxgr wrote: post #212

saagarjha wrote: post #201

Earlier quoted context omitted.

Secure PIN entry

How? The secure enclave doesn't have any trusted input/output capabilities other than the power button double click and biometrics. The PIN isn't entered via these.

There’s a thing which I know very little about personally but have hearsay of its existence where it rips control from XNU and takes over touch input for this

Re: Apple Exclaves

#233

Post by saagarjha » Sat, Mar 15, 2025, 10:40 AM UTC

kennysoona wrote: post #228

saagarjha wrote: post #89

Earlier quoted context omitted.

> In macOS 15 and later, creation of a VM running macOS 15 or later can configure an identity derived from the host Secure Enclave, enabling access to resources requiring Apple ID including iCloud. This is accomplished using an exclave of the Secure Enclave. This is not correct

How is it incorrect?

It just straight up doesn’t make sense. Honestly I would not be surprised if this was the result of going, hmm, enclave…exclave…seems like these must be related. They’re not really related other than the idea being isolation (which they do completely differently). Also Apple ID sign in on VMs is literally just “the host shares its information to the guest” there’s nothing special going on here