On Wed, Nov 29, 2023 at 04:04:03PM -0800, Andrii Nakryiko wrote:
> Given we enforce a valid range for program and async callback return
> value, we must mark R0 as precise to avoid incorrect state pruning.
Looking at previous discussion[1], this commit fixes the potential
"out-of-range r0 got state pruned" issue. To my best knowledge that
means this commit would be needed all the way back in
Fixes: b5dc0163d8fd ("bpf: precise scalar_value tracking")
Is this wildly off?
1: https://lore.kernel.org/r/20231031050324.1107444-4-andrii@xxxxxxxxxx
> Acked-by: Eduard Zingerman <eddyz87@xxxxxxxxx>
> Signed-off-by: Andrii Nakryiko <andrii@xxxxxxxxxx>
> ---
> kernel/bpf/verifier.c | 4 ++++
> 1 file changed, 4 insertions(+)
>
> diff --git a/kernel/bpf/verifier.c b/kernel/bpf/verifier.c
> index c54944af1bcc..2cd150d6d141 100644
> --- a/kernel/bpf/verifier.c
> +++ b/kernel/bpf/verifier.c
> @@ -15138,6 +15138,10 @@ static int check_return_code(struct bpf_verifier_env *env, int regno, const char
> return -EINVAL;
> }
>
> + err = mark_chain_precision(env, regno);
> + if (err)
> + return err;
> +
> if (!retval_range_within(range, reg)) {
> verbose_invalid_scalar(env, reg, range, exit_ctx, reg_name);
> if (!is_subprog &&
