On Wed, Mar 29, 2023 at 12:56 AM Andrii Nakryiko <andrii@xxxxxxxxxx> wrote:
>
> It's not clear why we have 128 as minimum size, but it makes testing
> harder and seems unnecessary, as we carefully handle truncation
> scenarios and use proper snprintf variants. So remove this limitation
> and just enfore positive length for log buffer.
Nit: enforce
>
> Signed-off-by: Andrii Nakryiko <andrii@xxxxxxxxxx>
Acked-by: Lorenz Bauer <lmb@xxxxxxxxxxxxx>
> ---
> kernel/bpf/log.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/kernel/bpf/log.c b/kernel/bpf/log.c
> index 920061e38d2e..1974891fc324 100644
> --- a/kernel/bpf/log.c
> +++ b/kernel/bpf/log.c
> @@ -11,7 +11,7 @@
>
> bool bpf_verifier_log_attr_valid(const struct bpf_verifier_log *log)
> {
> - return log->len_total >= 128 && log->len_total <= UINT_MAX >> 2 &&
> + return log->len_total > 0 && log->len_total <= UINT_MAX >> 2 &&
> log->level && log->ubuf && !(log->level & ~BPF_LOG_MASK);
Probably discussion for your second series. Thought experiment, could
this be len_total >= 0? I'm still after a way to get the correct log
buffer size from the first PROG_LOAD call. If the kernel could handle
this I could mmap a PROT_NONE page, set len_total to 0 and later read
out the correct buffer size.
I'm guessing that the null termination logic would have to be
adjusted, anything else though?
