[dpdk-dev] [PATCH] kni: unregister an unregisterd net_device could cause a kernel crash

Thomas Monjalon thomas.monjalon at 6wind.com
Sat Sep 10 09:57:20 CEST 2016


2016-09-09 14:33, Mcnamara, John:
> From: dev [mailto:dev-bounces at dpdk.org] On Behalf Of Thomas Monjalon
> > 2016-09-08 18:15, Ferruh Yigit:
> > > On 9/8/2016 5:44 PM, Stephen Hemminger wrote:
> > >
> > > ...
> > >
> > > > But then again the whole KNI driver fails completely when running
> > > > kernel style check.
> > > >
> > >
> > > Yes, it generates lots of warnings.
> > > I can fix them (excluding ethtool/*), that wouldn't take much time but
> > > how syntax only patches welcomed? Another concern is it trashes git
> > blame.
> > 
> > You ask a question and give the answer ;) I think it depends just on the
> > balance of the pros/cons - to be evaluated.
> 
> Hi,
> 
> I think in general we would prefer to avoid any large scale code beautification since, as pointed out, it breaks the option to git blame.
> 
> However, in the case of the KNI code the main author in git is "Intel" so git blame doesn't tell you a lot. Also, Ferruh is the maintainer, has made most of the recent changes, and is actively maintaining/improving it. So I think if the syntax fix came from him it would be okay. At least it would allow us to apply the checkpatch checks.

Yes seems reasonnable


More information about the dev mailing list