Information Exposure through Microarchitectural State after Transient Execution
The processor does not properly clear microarchitectural state after incorrect microcode assists or speculative execution, resulting in transient execution.
In many processor architectures an exception, mis-speculation, or microcode assist results in a flush operation to clear results that are no longer required. This action prevents these results from influencing architectural state that is intended to be visible from software. However, traces of this transient execution may remain in microarchitectural buffers, resulting in a change in microarchitectural state that can expose sensitive information to an attacker using side-channel analysis. For example, Load Value Injection (LVI) [REF-1202] can exploit direct injection of erroneous values into intermediate load and store buffers.
Several conditions may need to be fulfilled for a successful attack:
) incorrect transient execution that results in remanence of sensitive information;
) attacker has the ability to provoke microarchitectural exceptions;
) operations and structures in victim code that can be exploited must be identified.
The following examples help to illustrate the nature of this weakness and describe methods or techniques which can be used to mitigate the risk.
Note that the examples here are by no means exhaustive and any given weakness may have many subtle varieties, each of which may require different detection methods or runtime controls.
Faulting loads in a victim domain may trigger incorrect transient forwarding, which leaves secret-dependent traces in the microarchitectural state. Consider this example from [REF-1203].
Consider the code gadget:
A processor with this weakness will store the value of untrusted_arg (which may be provided by an attacker) to the stack, which is trusted memory. Additionally, this store operation will save this value in some microarchitectural buffer, e.g. the store queue.
In this code gadget, trusted_ptr is dereferenced while the attacker forces a page fault. The faulting load causes the processor to mis-speculate by forwarding untrusted_arg as the (speculative) load result. The processor then uses untrusted_arg for the pointer dereference. After the fault has been handled and the load has been re-issued with the correct argument, secret-dependent information stored at the address of trusted_ptr remains in microarchitectural state and can be extracted by an attacker using a code gadget.
Weaknesses in this category are typically associated with CPUs, Graphics, Vision, AI, FPGA, and microcontrollers.
This view (slice) covers all the elements in CWE.
CWE identifiers in this view are weaknesses that do not have associated Software Fault Patterns (SFPs), as covered by the CWE-888 view. As such, they represent gaps in...
This view (slice) lists weaknesses that can be introduced during design.