Thanks, I will provide a new patch

> -----Original Message-----
> From: [email protected] [mailto:[email protected]]
> On Behalf Of David Miller
> Sent: Wednesday, November 15, 2017 1:20 PM
> To: Yuan, Linyu (NSB - CN/Shanghai)
> Cc: [email protected]; [email protected]
> Subject: Re: [PATCH net-next] net: assign err to 0 at begin in do_setlink()
> function
> 
> From: "Yuan, Linyu (NSB - CN/Shanghai)" <[email protected]>
> Date: Wed, 15 Nov 2017 05:15:35 +0000
> 
> >
> >
> >> -----Original Message-----
> >> From: [email protected]
> [mailto:[email protected]]
> >> On Behalf Of David Miller
> >> Sent: Wednesday, November 15, 2017 1:08 PM
> >> To: [email protected]
> >> Cc: [email protected]; Yuan, Linyu (NSB - CN/Shanghai)
> >> Subject: Re: [PATCH net-next] net: assign err to 0 at begin in do_setlink()
> >> function
> >>
> >> From: yuan linyu <[email protected]>
> >> Date: Tue, 14 Nov 2017 22:30:59 +0800
> >>
> >> > From: yuan linyu <[email protected]>
> >> >
> >> > each netlink attribute have proper process when error happen,
> >> > when exit on attribute process, it implies that no error,
> >> > so err = 0; is useless.
> >> >
> >> > assign err = 0; at beginning if all attributes not set.
> >> >
> >> > Signed-off-by: yuan linyu <[email protected]>
> >>
> >> This change is not correct.
> >>
> >> The IFLA_VF_PORTS code block can finish with err set non-zero.
> >
> > Do you mean this block can return err = -EOPNOTSUPP;  ?
> > It means nla_for_each_nested() in this block will never enter, right ?
> 
> Yes, that is what I mean.

Reply via email to