On 12/11/2018 03:03 AM, David Miller wrote: > From: Daniel Borkmann <dan...@iogearbox.net> > Date: Tue, 11 Dec 2018 01:33:56 +0100 > >> It has three minor merge conflicts, resolutions: >> >> 1) tools/testing/selftests/bpf/test_verifier.c >> >> Take first chunk with alignment_prevented_execution. >> >> 2) net/core/filter.c >> >> [...] >> case bpf_ctx_range_ptr(struct __sk_buff, flow_keys): >> case bpf_ctx_range(struct __sk_buff, wire_len): >> return false; >> [...] >> >> 3) include/uapi/linux/bpf.h >> >> Take the second chunk for the two cases each. > > Thanks for this guidance. > > I'm not %100 sure I got case #3 correct. The two sets of > text talk about hashing over the "packet" vs. the "tuple". > These intefaces take a tuple, so it only makes sense to > talk about hashing over a tuple so I chose the hunk > which says "tuple". > > I had to deal with this during the net --> net-next merge > last night as well.
Yes, absolutely, that is correct. > Please double check my work and send me any relative fixups which > might be necessary. I've sent a minor fixup here to add back the formatting improvements: http://patchwork.ozlabs.org/patch/1010914/ Thanks, Daniel
