Hello,
On Fri, 19 Apr 2013, Simon Horman wrote:
> My idea is as follows:
>
> From: Simon Horman <horms@xxxxxxxxxxxx>
>
> ipvs: Use network byte order for sync message size
>
> struct ip_vs_sync_mesg is both sent across the wire and
> used internally to store IPVS synchronisation messages.
>
> Up until now the scheme used has been to convert the size field
> to network byte order before sending a message on the wire and
> convert it to host byte order when sending a message.
>
> This patch changes that scheme to always treat the field
> as being network byte order. This seems appropriate as
> the structure is sent across the wire. And by consistently
> treating the field has network byte order it is now possible
> to take advantage of sparse to flag any future miss-use.
>
> Signed-off-by: Simon Horman <horms@xxxxxxxxxxxx>
> ---
> net/netfilter/ipvs/ip_vs_sync.c | 15 +++++----------
> 1 file changed, 5 insertions(+), 10 deletions(-)
>
> diff --git a/net/netfilter/ipvs/ip_vs_sync.c b/net/netfilter/ipvs/ip_vs_sync.c
> index 8e57077..c73778a 100644
> --- a/net/netfilter/ipvs/ip_vs_sync.c
> +++ b/net/netfilter/ipvs/ip_vs_sync.c
> @@ -255,7 +255,7 @@ struct ip_vs_sync_mesg_v0 {
> struct ip_vs_sync_mesg {
> __u8 reserved; /* must be zero */
> __u8 syncid;
> - __u16 size;
> + __be16 size;
It seems v0 should be changed too,
ip_vs_send_sync_msg() handles both versions in sb->mesg.
> @@ -1544,10 +1542,7 @@ ip_vs_send_sync_msg(struct socket *sock, struct ip_vs_sync_mesg *msg)
> int msize;
> int ret;
>
> - msize = msg->size;
> -
> - /* Put size in network byte order */
> - msg->size = htons(msg->size);
> + msize = ntohs(msg->size);
>
> ret = ip_vs_send_async(sock, (char *)msg, msize);
> if (ret >= 0 || ret == -EAGAIN)
Regards
--
Julian Anastasov <ja@xxxxxx>
--
To unsubscribe from this list: send the line "unsubscribe lvs-devel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
