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: <7dcee85b-b3ef-947f-f433-03ad7066c5dd@nvidia.com>
Date:   Wed, 20 Jan 2021 15:15:22 +0200
From:   Maxim Mikityanskiy <maximmi@...dia.com>
To:     Björn Töpel <bjorn.topel@...il.com>,
        <ast@...nel.org>, <daniel@...earbox.net>, <netdev@...r.kernel.org>,
        <bpf@...r.kernel.org>
CC:     <bjorn.topel@...el.com>, <magnus.karlsson@...el.com>,
        <maciej.fijalkowski@...el.com>, <kuba@...nel.org>,
        <jonathan.lemon@...il.com>, <davem@...emloft.net>,
        <hawk@...nel.org>, <john.fastabend@...il.com>,
        <ciara.loftus@...el.com>, <weqaar.a.janjua@...el.com>
Subject: Re: [PATCH bpf-next v2 0/8] Introduce bpf_redirect_xsk() helper

On 2021-01-19 17:50, Björn Töpel wrote:
> This series extends bind() for XDP sockets, so that the bound socket
> is added to the netdev_rx_queue _rx array in the netdevice. We call
> this to register the socket. To redirect packets to the registered
> socket, a new BPF helper is used: bpf_redirect_xsk().
> 
> For shared XDP sockets, only the first bound socket is
> registered. Users that need more complex setup has to use XSKMAP and
> bpf_redirect_map().
> 
> Now, why would one use bpf_redirect_xsk() over the regular
> bpf_redirect_map() helper?
> 
> * Better performance!
> * Convenience; Most user use one socket per queue. This scenario is
>    what registered sockets support. There is no need to create an
>    XSKMAP. This can also reduce complexity from containerized setups,
>    where users might what to use XDP sockets without CAP_SYS_ADMIN
>    capabilities.
> 
> The first patch restructures xdp_do_redirect() a bit, to make it
> easier to add the new helper. This restructure also give us a slight
> performance benefit. The following three patches extends bind() and
> adds the new helper. After that, two libbpf patches that selects XDP
> program based on what kernel is running. Finally, selftests for the new
> functionality is added.
> 
> Note that the libbpf "auto-selection" is based on kernel version, so
> it is hard coded to the "-next" version (5.12). If you would like to
> try this is out, you will need to change the libbpf patch locally!
> 
> Thanks to Maciej and Magnus for the internal review/comments!
> 
> Performance (rxdrop, zero-copy)
> 
> Baseline
> Two cores:                   21.3 Mpps
> One core:                    24.5 Mpps

Two cores is slower? It used to be faster all the time, didn't it?

> Patched
> Two cores, bpf_redirect_map: 21.7 Mpps + 2%
> One core, bpf_redirect_map:  24.9 Mpps + 2%
> 
> Two cores, bpf_redirect_xsk: 24.0 Mpps +13%

Nice, impressive improvement!

> One core, bpf_redirect_xsk:  25.5 Mpps + 4%
> 
> Thanks!
> Björn
> 
> v1->v2:
>    * Added missing XDP programs to selftests.
>    * Fixed checkpatch warning in selftests.
> 
> Björn Töpel (8):
>    xdp: restructure redirect actions
>    xsk: remove explicit_free parameter from __xsk_rcv()
>    xsk: fold xp_assign_dev and __xp_assign_dev
>    xsk: register XDP sockets at bind(), and add new AF_XDP BPF helper
>    libbpf, xsk: select AF_XDP BPF program based on kernel version
>    libbpf, xsk: select bpf_redirect_xsk(), if supported
>    selftest/bpf: add XDP socket tests for bpf_redirect_{xsk, map}()
>    selftest/bpf: remove a lot of ifobject casting in xdpxceiver
> 
>   include/linux/filter.h                        |  10 +
>   include/linux/netdevice.h                     |   1 +
>   include/net/xdp_sock.h                        |  12 +
>   include/net/xsk_buff_pool.h                   |   2 +-
>   include/trace/events/xdp.h                    |  46 ++--
>   include/uapi/linux/bpf.h                      |   7 +
>   net/core/filter.c                             | 205 ++++++++++--------
>   net/xdp/xsk.c                                 | 112 ++++++++--
>   net/xdp/xsk_buff_pool.c                       |  12 +-
>   tools/include/uapi/linux/bpf.h                |   7 +
>   tools/lib/bpf/libbpf.c                        |   2 +-
>   tools/lib/bpf/libbpf_internal.h               |   2 +
>   tools/lib/bpf/libbpf_probes.c                 |  16 --
>   tools/lib/bpf/xsk.c                           |  83 ++++++-
>   .../selftests/bpf/progs/xdpxceiver_ext1.c     |  15 ++
>   .../selftests/bpf/progs/xdpxceiver_ext2.c     |   9 +
>   tools/testing/selftests/bpf/test_xsk.sh       |  48 ++++
>   tools/testing/selftests/bpf/xdpxceiver.c      | 164 +++++++++-----
>   tools/testing/selftests/bpf/xdpxceiver.h      |   2 +
>   19 files changed, 554 insertions(+), 201 deletions(-)
>   create mode 100644 tools/testing/selftests/bpf/progs/xdpxceiver_ext1.c
>   create mode 100644 tools/testing/selftests/bpf/progs/xdpxceiver_ext2.c
> 
> 
> base-commit: 95204c9bfa48d2f4d3bab7df55c1cc823957ff81
> 

Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ