> Vladimir Medvedkin, Jul 18, 2024 at 23:25: > > I think alignment should be 1 since in FIB6 users usually don't copy IPv6 > > address and just provide a pointer to the memory inside the packet. Current > > vector implementation loads IPv6 addresses using unaligned access ( > > _mm512_loadu_si512) so it doesn't rely on alignment. > > Yes, my intention was exactly that, being able to map that structure > directly in packets without copying them on the stack. > > > > 2. In the IPv6 packet header, the IPv6 addresses are not 16 byte aligned, > > > they are 8 byte aligned. So we cannot make the IPv6 address type 16 byte > > > aligned. > > > Not necessary, if Ethernet frame in mbuf starts on 8b aligned address, then > > IPv6 is aligned only by 2 bytes. > > We probably could safely say that aligning on 2 bytes would be OK. But > is there any benefit, performance wise, in doing so? Keeping the same > alignment as before the change would at least make it ABI compatible.
I am also not sure that this extra alignment (2B or 4B) here will give us any benefit, while it most likely will introduce extra restrictions. AFAIK, right now we do have ipv6 as array of plain chars, and there were no much complaints about it. So I am for keeping it 1B aligned. Overall proposal looks reasonable to me... might be 24.11 is a good opportunity for such change. Konstantin
