On Sun, 27 Nov 2016 16:51:12 +0200, Yuval Mintz wrote:
> Add support for the ndo_xdp callback. This patch would support XDP_PASS,
> XDP_DROP and XDP_ABORTED commands.
>
> This also adds a per Rx queue statistic which counts number of packets
> which didn't reach the stack [due to XDP].
>
> Signed-off-by: Yuval Mintz <yuval.mi...@cavium.com>
> ---
> [...]
> +static int qede_xdp_set(struct qede_dev *edev, struct bpf_prog *prog)
> +{
> + bool reload_required = true;
> + struct qede_reload_args args;
> + int rc = 0;
> +
> + /* Protect against various other internal-reload flows */
> + __qede_lock(edev);
> + if (edev->state != QEDE_STATE_OPEN) {
> + rc = -EINVAL;
> + goto out;
> + }
> +
> + /* If we're called, there was already a bpf reference increment */
> + args.func = &qede_xdp_reload_func;
> + args.u.new_prog = prog;
> + if (reload_required)
> + qede_reload(edev, &args, true);
> + else
> + args.func(edev, &args);
> +
> +out:
> + __qede_unlock(edev);
> + return rc;
> +}
Any particular reason not to allow the XDP prog being set while the
device is closed? You seem to preserve the program across
close()/open() cycles so the edev->xdp_prog is alive and valid while
device is closed IIUC.
I think other drivers are allowing setting XDP while closed and it
would be cool to keep the behaviour identical across drivers :)
