In the Linux kernel, the following vulnerability has been resolved: bpf: account for current allocated stack depth in widen_imprecise_scalars() The usage pattern for widen_imprecise_scalars() looks as follows: prev_st = find_prev_entry(env, ...); queued_st = push_stack(...); widen_imprecise_scalars(env, prev_st, queued_st); Where prev_st is an ancestor of the queued_st in the explored states tree. This ancestor is not guaranteed to have same allocated stack depth as queued_st. E.g. in the following case: def main(): for i in 1..2: foo(i) // same callsite, differnt param def foo(i): if i == 1: use 128 bytes of stack iterator based loop Here, for a second 'foo' call prev_st->allocated_stack is 128, while queued_st->allocated_stack is much smaller. widen_imprecise_scalars() needs to take this into account and avoid accessing bpf_verifier_state->frame[*]->stack out of bounds.
https://git.kernel.org/stable/c/b0c8e6d3d866b6a7f73877f71968dbffd27b7785
https://git.kernel.org/stable/c/9944c7938cd5b3f37b0afec0481c7c015e4f1c58
https://git.kernel.org/stable/c/64b12dca2b0abcb5fc0542887d18b926ea5cf711
https://git.kernel.org/stable/c/57e04e2ff56e32f923154f0f7bc476fcb596ffe7