Re: [PATCH bpf-next v2 5/8] libbpf, xsk: select AF_XDP BPF program based on kernel version

Björn Töpel [off-list ref] writes:

On 2021-01-20 13:52, Toke Høiland-Jørgensen wrote:

quoted

Björn Töpel [off-list ref] writes:

quoted

From: Björn Töpel <redacted>

Add detection for kernel version, and adapt the BPF program based on
kernel support. This way, users will get the best possible performance
from the BPF program.

Please do explicit feature detection instead of relying on the kernel
version number; some distro kernels are known to have a creative notion
of their own version, which is not really related to the features they
actually support (I'm sure you know which one I'm referring to ;)).

Right. For a *new* helper, like bpf_redirect_xsk, we rely on rejection
from the verifier to detect support. What about "bpf_redirect_map() now
supports passing return value as flags"? Any ideas how to do that in a
robust, non-version number-based scheme?

Well, having a BPF program pass in a flag of '1' with an invalid lookup
and checking if it returns 1 or 0. But how to do that from libbpf, hmm,
good question. BPF_PROG_RUN()?

An alternative could be to default to a program that will handle both
cases in the BPF code, and make it opt-in to use the optimised versions
if the user knows their kernel supports it?

-Toke