Messages in this thread | | | From | Jay Vosburgh <> | Subject | Re: [PATCH 1/2] net: bonding: fix restricted __be16 degrades to integer | Date | Fri, 08 Mar 2019 12:54:46 -0800 |
| |
Bo YU <tsu.yubo@gmail.com> wrote:
>There are some warning when: > >sudo make C=1 CF=-D__CHECK_ENDIAN__ drivers/net/bonding/ > >drivers/net/bonding/bond_main.c:2385:26: warning: restricted __be16 degrades to integer >drivers/net/bonding/bond_main.c:2391:20: warning: restricted __be16 degrades to integer >... >drivers/net/bonding/bond_main.c:3241:60: warning: restricted __be16 degrades to integer >drivers/net/bonding/bond_main.c:3241:60: warning: restricted __be16 degrades to integer > >So fix it. > >Signed-off-by: Bo YU <tsu.yubo@gmail.com> >--- > drivers/net/bonding/bond_main.c | 6 +++--- > 1 file changed, 3 insertions(+), 3 deletions(-) > >diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c >index b59708c35faf..135fec28daa9 100644 >--- a/drivers/net/bonding/bond_main.c >+++ b/drivers/net/bonding/bond_main.c >@@ -2382,13 +2382,13 @@ static void bond_arp_send(struct net_device *slave_dev, int arp_op, > return; > } > >- if (!tags || tags->vlan_proto == VLAN_N_VID) >+ if (!tags || be16_to_cpu(tags->vlan_proto) == VLAN_N_VID) > goto xmit; > > tags++; > > /* Go through all the tags backwards and add them to the packet */ >- while (tags->vlan_proto != VLAN_N_VID) { >+ while (be16_to_cpu(tags->vlan_proto) != VLAN_N_VID) {
I believe both of the above are incorrect, as vlan_proto is set explicitly to VLAN_N_VID (in host byte order) by bond_verify_device_path as a sentinel value. Byte swapping the tags->vlan_proto value would cause the test or loop to miss the sentinel.
> if (!tags->vlan_id) { > tags++; > continue; >@@ -3238,7 +3238,7 @@ static inline u32 bond_eth_hash(struct sk_buff *skb) > > ep = skb_header_pointer(skb, 0, sizeof(hdr_tmp), &hdr_tmp); > if (ep) >- return ep->h_dest[5] ^ ep->h_source[5] ^ ep->h_proto; >+ return ep->h_dest[5] ^ ep->h_source[5] ^ be16_to_cpu(ep->h_proto);
This is probably harmless, other than adding work to the transmit path.
-J
--- -Jay Vosburgh, jay.vosburgh@canonical.com
| |