With common OSes (e.g. Linux), threads are scheduled based on timer interrupts. But SGX is somewhat different since interrupts are handled as AEXes. So I'm wondering are SGX threads subject to OS scheduling too? Is there a way to mask timer interrupts so that a SGX thread can continue execution to the end without being scheduled out?
From an OS perspective, enclave threads are just normal threads. The OS does not “know” about SGX. So, SGX threads are treated the same way as normal threads and can be scheduled out like any other thread.
From an OS perspective, enclave threads are just normal threads. The OS does not “know” about SGX. So, SGX threads are treated the same way as normal threads and can be scheduled out like any other thread.
