netdev - Re: Resource management for ndo_xdp_xmit (Was: [PATCH net] virtio_net: Account for tx bytes and packets on sending xdp

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

Message-ID: <87d0o0q4t1.fsf@toke.dk>
Date:   Sat, 09 Feb 2019 17:56:10 +0100
From:   Toke Høiland-Jørgensen <toke@...hat.com>
To:     Saeed Mahameed <saeedm@...lanox.com>,
        "brouer\@redhat.com" <brouer@...hat.com>
Cc:     "hawk\@kernel.org" <hawk@...nel.org>,
        "virtualization\@lists.linux-foundation.org" 
        <virtualization@...ts.linux-foundation.org>,
        "borkmann\@iogearbox.net" <borkmann@...earbox.net>,
        Tariq Toukan <tariqt@...lanox.com>,
        "john.fastabend\@gmail.com" <john.fastabend@...il.com>,
        "mst\@redhat.com" <mst@...hat.com>,
        "jakub.kicinski\@netronome.com" <jakub.kicinski@...ronome.com>,
        "dsahern\@gmail.com" <dsahern@...il.com>,
        "netdev\@vger.kernel.org" <netdev@...r.kernel.org>,
        "jasowang\@redhat.com" <jasowang@...hat.com>,
        "davem\@davemloft.net" <davem@...emloft.net>,
        "makita.toshiaki\@lab.ntt.co.jp" <makita.toshiaki@....ntt.co.jp>
Subject: Re: Resource management for ndo_xdp_xmit (Was: [PATCH net] virtio_net: Account for tx bytes and packets on sending xdp_frames)

Saeed Mahameed <saeedm@...lanox.com> writes:

> On Fri, 2019-02-08 at 15:17 -0800, Saeed Mahameed wrote:
>> On Thu, 2019-02-07 at 19:08 +0000, Saeed Mahameed wrote:
>> > 
>> > So 
>> > 1) on dev_map_update_elem() we will call
>> > dev->dev->ndo_bpf() to notify the device on the intention to
>> > start/stop
>> > redirect, and wait for it to create/destroy the HW resources
>> > before/after actually updating the map
>> > 
>> 
>> silly me, dev_map_update_elem must be atomic, we can't hook driver
>> resource allocation to it, it must come as a separate request
>> (syscall)
>> from user space to request to create XDP redirect resources.
>> 
>
> Well, it is possible to render dev_map_update_elem non-atomic and fail
> BPF programs who try to update it in the verifier
> check_map_func_compatibility.
>
> if you know of any case where devmap needs to be updated from the BPF
> program please let me know.

Well, maybe? My plan for dealing with the non-map redirect variant is
essentially to have a hidden map that does just-in-time insertion of
ifindexes if they are not already there; and rework XDP_REDIRECT to use
that.

However, this would essentially be an insert from eBPF; but I guess
maybe it could be deferred until the RX-side driver exits its NAPI loop
(as we're buffering frames in the map anyway)?

-Toke