A major and the most significant approach to UEFI BIOS security is to prevent it from being illegitimately modified and the SPI flash memory from being overwritten. Modern vendors use a wide range of security mechanisms to ensure that (SMM BLE / SMM BWP / PRx / Intel BIOS Guard) and hardware-supported verification technologies (Intel Boot Guard). In other words, they do everything just not to let an attacker place a rootkit into a system.
Even the likelihood of execution in the most privileged mode of a processor – System Management Mode (can be achieved through vulnerable software SMI handlers) – is of no interest to adversaries since it does not guarantee they will be able to gain a foothold in a system. A single reboot and an attack must be started anew.
However, there is a thing that can make all BIOS security mechanisms inefficient. And this thing is a vulnerable update mechanism implemented by a vendor. Moreover, quite often a legitimate updater adds lots and lots of critical security holes to a system.
In this talk, we will speak about how vendors manage to throw all those security flaws together in one system using Intel NUC, a small home PC, as an example. Besides, we will demonstrate how an adversary can compromise BIOS from the userland.