netdev - RE: [PATCH net-next v7 1/2] macsec: add support for IFLA_MACSEC_OFFLOAD in macsec

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Message-ID: <IA1PR12MB635353E467A1BE267076D269ABFF9@IA1PR12MB6353.namprd12.prod.outlook.com>
Date:   Tue, 10 Jan 2023 16:16:43 +0000
From:   Emeel Hakim <ehakim@...dia.com>
To:     Antoine Tenart <atenart@...nel.org>,
        Sabrina Dubroca <sd@...asysnail.net>
CC:     "netdev@...r.kernel.org" <netdev@...r.kernel.org>,
        Raed Salem <raeds@...dia.com>,
        "davem@...emloft.net" <davem@...emloft.net>,
        "edumazet@...gle.com" <edumazet@...gle.com>,
        "kuba@...nel.org" <kuba@...nel.org>,
        "pabeni@...hat.com" <pabeni@...hat.com>
Subject: RE: [PATCH net-next v7 1/2] macsec: add support for
 IFLA_MACSEC_OFFLOAD in macsec_changelink



> -----Original Message-----
> From: Antoine Tenart <atenart@...nel.org>
> Sent: Tuesday, 10 January 2023 15:55
> To: Sabrina Dubroca <sd@...asysnail.net>
> Cc: Emeel Hakim <ehakim@...dia.com>; netdev@...r.kernel.org; Raed Salem
> <raeds@...dia.com>; davem@...emloft.net; edumazet@...gle.com;
> kuba@...nel.org; pabeni@...hat.com
> Subject: Re: [PATCH net-next v7 1/2] macsec: add support for
> IFLA_MACSEC_OFFLOAD in macsec_changelink
> 
> External email: Use caution opening links or attachments
> 
> 
> Quoting Sabrina Dubroca (2023-01-10 11:44:13)
> > 2023-01-10, 09:43:37 +0100, Antoine Tenart wrote:
> > > Quoting Sabrina Dubroca (2023-01-09 16:14:32)
> > > > 2023-01-09, 10:55:56 +0200, ehakim@...dia.com wrote:
> > > > > @@ -3840,6 +3835,12 @@ static int macsec_changelink(struct net_device
> *dev, struct nlattr *tb[],
> > > > >       if (ret)
> > > > >               goto cleanup;
> > > > >
> > > > > +     if (data[IFLA_MACSEC_OFFLOAD]) {
> > > > > +             ret = macsec_update_offload(dev,
> nla_get_u8(data[IFLA_MACSEC_OFFLOAD]));
> > > > > +             if (ret)
> > > > > +                     goto cleanup;
> > > > > +     }
> > > > > +
> > > > >       /* If h/w offloading is available, propagate to the device */
> > > > >       if (macsec_is_offloaded(macsec)) {
> > > > >               const struct macsec_ops *ops;
> > > >
> > > > There's a missing rollback of the offloading status in the
> > > > (probably quite unlikely) case that mdo_upd_secy fails, no? We
> > > > can't fail macsec_get_ops because macsec_update_offload would have
> > > > failed already, but I guess the driver could fail in mdo_upd_secy,
> > > > and then "goto cleanup" doesn't restore the offloading state.
> > > > Sorry I didn't notice this earlier.
> > > >
> > > > In case the IFLA_MACSEC_OFFLOAD attribute is provided and we're
> > > > enabling offload, we also end up calling the driver's
> > > > mdo_add_secy, and then immediately afterwards mdo_upd_secy, which
> > > > probably doesn't make much sense.
> > > >
> > > > Maybe we could turn that into:
> > > >
> > > >     if (data[IFLA_MACSEC_OFFLOAD]) {
> > >
> > > If data[IFLA_MACSEC_OFFLOAD] is provided but doesn't change the
> > > offloading state, then macsec_update_offload will return early and
> > > mdo_upd_secy won't be called.
> >
> > Ouch, thanks for catching this.
> >
> > >
> > > >         ... macsec_update_offload
> > > >     } else if (macsec_is_offloaded(macsec)) {
> > > >         /* If h/w offloading is available, propagate to the device */
> > > >         ... mdo_upd_secy
> > > >     }
> > > >
> > > > Antoine, does that look reasonable to you?
> > >
> > > But yes I agree we can improve the logic. Maybe something like:
> > >
> > >   prev_offload = macsec->offload;
> > >   offload = data[IFLA_MACSEC_OFFLOAD];
> >
> > That needs to be under if (data[IFLA_MACSEC_OFFLOAD]) and then the
> > rest gets a bit messy.
> >
> > >
> > >   if (prev_offload != offload) {
> > >       macsec_update_offload(...)
> > >   } else if (macsec_is_offloaded(macsec)) {
> > >       ...
> > >       prev_offload can be used to restore the offloading state on
> > >       failure here.
> > >   }
> >
> > We also have a prev != new test at the start of macsec_update_offload,
> > the duplication is a bit ugly. We could move it out and then only call
> > macsec_update_offload when there is a change to do, both from
> > macsec_changelink and macsec_upd_offload.
> 
> Agreed.
> 
> > Since we don't need to restore in the second branch, and we can only
> > fetch IFLA_MACSEC_OFFLOAD when it's present, maybe:
> >
> >     change = false;
> >     if (data[IFLA_MACSEC_OFFLOAD]) {
> >         offload = nla_get_u8(data[IFLA_MACSEC_OFFLOAD]);
> >         if (macsec->offload != offload) {
> >             change = true;
> >             macsec_update_offload ...cleanup
> >         }
> >     }
> >
> >     if (!change && macsec_is_offloaded(macsec)) {
> >         ...
> >     }
> >
> > Or let macsec_update_offload do the macsec->offload != offload test
> > and pass &change so that changelink can know what to do next.
> 
> Either solutions work for me.

Ack, I will send a v8 with Sabrina's approach of change = false ...

> Thanks!
> Antoine

Thanks,
Emeel