On Wed, Apr 26, 2023 at 06:39:22PM +0200, Ilya Maximets wrote:
> On 4/25/23 14:41, Roi Dayan wrote:
> > From: Gavin Li <[email protected]>
> > 
> > Linux kernel netlink module added NLA_F_NESTED flag checking for nested
> > netlink messages in 5.2. A nested message without the flag set will be
> > treated as malformated one. The check is optional and is controlled by
> > message policy. To avoid this, add NLA_F_NESTED explicitly for all
> > nested netlink messages.
> > 
> > Signed-off-by: Gavin Li <[email protected]>
> > Reviewed-by: Roi Dayan <[email protected]>
> > ---
> >  lib/netlink.c | 2 +-
> >  1 file changed, 1 insertion(+), 1 deletion(-)
> > 
> > diff --git a/lib/netlink.c b/lib/netlink.c
> > index 6215282d6fbe..f128b63074f9 100644
> > --- a/lib/netlink.c
> > +++ b/lib/netlink.c
> > @@ -519,7 +519,7 @@ size_t
> >  nl_msg_start_nested(struct ofpbuf *msg, uint16_t type)
> >  {
> >      size_t offset = msg->size;
> > -    nl_msg_put_unspec_uninit(msg, type, 0);
> > +    nl_msg_put_unspec_uninit(msg, type | NLA_F_NESTED, 0);
> >      return offset;
> >  }
> >  
> 
> 
> Just posting a link to the ongoing v1 conversation on this patch here
> for visibility:
>   
> https://patchwork.ozlabs.org/project/openvswitch/patch/[email protected]/

It seems to me that the agreement in the thread at the link above was to
create a function, nl_msg_start_nested_with_flag(), that would only set the
flag when appropriate. Which I think was the TC use-case.

Did the conversation move on from there?
Or was it an oversight not to do that in v2?
Or am I completely off track?
_______________________________________________
dev mailing list
[email protected]
https://mail.openvswitch.org/mailman/listinfo/ovs-dev

Reply via email to