Re: [PATCH v3 2/2] mac80211: allow rx of mesh eapol frames with default rx key

[Johannes Berg <johannes@xxxxxxxxxxxxxxxx>]

> +	/* check mesh EAPOL frames first */
> +	if (unlikely(rx->sta && ieee80211_vif_is_mesh(&rx->sdata->vif) && ieee80211_is_data(fc))) {
> +		struct ieee80211s_hdr *mesh_hdr;
> +		u16 hdr_len = ieee80211_hdrlen(fc);
> +		u16 ethertype_offset;
> +		__be16 ethertype;
> +
> +		/* make sure fixed part of mesh header is there, also checks skb len */
> +		if (!pskb_may_pull(rx->skb, hdr_len + 6))
> +			goto drop_check;
> +
> +		mesh_hdr = (struct ieee80211s_hdr *)(skb->data + hdr_len);
> +		ethertype_offset = hdr_len + ieee80211_get_mesh_hdrlen(mesh_hdr)
> +				 + sizeof(rfc1042_header);
> +		if (!pskb_may_pull(rx->skb, ethertype_offset + sizeof(ethertype)) ||
> +		    !ether_addr_equal(hdr->addr1, rx->sdata->vif.addr))

might be nicer to check the address first, pskb_may_pull() is
potentially more expensive, and you should be able to check the header
address already before even the first pskb_may_pull() here since it's in
the normal header.

But I don't understand the second pskb_may_pull() anyway, because you
use skb_copy_bits() below?


> +		skb_copy_bits(rx->skb, ethertype_offset, &ethertype, 2);
> +		if (ethertype == rx->sdata->control_port_protocol)
> +			goto pass_frame;

can return 0 here immediately and save the label.

johannes