On Tue, 14 Feb 2017 22:08:10 +0000 Edward Cree <ec...@solarflare.com> wrote:
> To me, the sensible interface (which makes the batching explicit to > the driver, which I think is necessary) is to have an int (or maybe > unsigned int, which is the return type of xdp_hookfn, I'm not sure > which is intended) member in struct xdp_buff. > > Then the driver can call something like > XDP_RUN_ARRAY(napi, xdp_array, array_len); > which is semantically equivalent to > unsigned int i; > for (i = 0; i < array_len; i++) > xdp_array[i].ret = xdp_hook_run(napi, xdp_array + i); Yes, exactly. I imagined the xdp_array[i].ret would be the XDP action return code. > except that it may run the hooks in 'row-major order'. > No callbacks needed, the driver can just loop over xdp_array reading > the .ret and applying the relevant action to each packet. > > This also has the advantage that the driver knows how many packets it > might have to process in a single batch (i.e. NAPI_POLL_WEIGHT) and > can allocate the array statically, whereas an XDP hook that tried to > transparently be 'helpful' would have to guess and/or use kmalloc. I also think the driver need to be explicit about batching. This related to the RX stages I'm talking about. Saeed is working on implementing that for mlx5, I got some PoC patches today and I'll soon test that. -- Best regards, Jesper Dangaard Brouer MSc.CS, Principal Kernel Engineer at Red Hat LinkedIn: http://www.linkedin.com/in/brouer
