netdev - Re: [PATCH net-next 10/11] qede: Add basic XDP support

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>] [day] [month] [year] [list]

Message-ID: <BL2PR07MB2306004B997234AEEDE2F44B8D8A0@BL2PR07MB2306.namprd07.prod.outlook.com>
Date:   Mon, 28 Nov 2016 20:53:29 +0000
From:   "Mintz, Yuval" <Yuval.Mintz@...ium.com>
To:     Tom Herbert <tom@...bertland.com>
CC:     "David S. Miller" <davem@...emloft.net>,
        Linux Kernel Network Developers <netdev@...r.kernel.org>
Subject: Re: [PATCH net-next 10/11] qede: Add basic XDP support

> > +       if (act == XDP_PASS)
> > +               return true;
> > +
> > +       /* Count number of packets not to be passed to stack */
> > +       rxq->xdp_no_pass++;
> > +
> > +       switch (act) {
> > +       default:
> > +               bpf_warn_invalid_xdp_action(act);

> XDP_TX is a valid action and in fact some drivers already support that.
> Given that this isn't the first instance of driver not supporting XDP_TX
> I think we need to clear define what means. Personally, I think that we
> shouldn't allow a program to load that  returns XDP_TX but driver does
> not support it. I believe Jesper might be looking into capabilities
> support for XDP to handle that. For the purposes of this patch I'd suggest
> having an XDP_TX case and warn user that an unsupported action
> as opposed to invalid  one was returned.

Patch #11 does add XDP_TX support.
Adding an explicit case with a warning to be removed in the next patch
sounds like a waste to me. But if you think 'future generations' would
benefit from it, sure.
 
> > +       case XDP_ABORTED:
> > +       case XDP_DROP: