On Mon, 2024-01-08 at 05:49 -0800, Maciej Żenczykowski wrote:
> (I've looked at all 3 patches, and they seem fine... but I don't claim
> to understand the intricacies of the verifier/registers/etc well
> enough to be certain)
Thank you.
> On Mon, Jan 8, 2024 at 5:28 AM Eduard Zingerman <eddyz87@xxxxxxxxx> wrote:
> >
> > Extend try_match_pkt_pointers() to handle == and != operations.
> > For instruction:
> >
> > .--------------- pointer to packet with some range R
> > | .--------- pointer to packet end
> > v v
> > if rA == rB goto ...
>
> it's possible this would be better without the 'goto' as just 'if (rA
> == rB) ...'
The idea was to show this as BPF asm instruction, which has syntax
'if rA == rB goto C'. I'll wait for more feedback and submit v2
with updated commit message to make it more clear.
> > It is valid to infer that R bytes are available in packet.
> > This change should allow verification of BPF generated for
> > C code like below:
> >
> > if (data + 42 != data_end) { ... }
>
> this should probably be:
> if (data + 42 != data_end) return;
> ...
Makes sense, will update commit message.
