summaryrefslogtreecommitdiffstats
path: root/Documentation
diff options
context:
space:
mode:
authorDavid S. Miller <davem@davemloft.net>2019-06-07 14:46:47 -0700
committerDavid S. Miller <davem@davemloft.net>2019-06-07 14:46:47 -0700
commit38e406f600a2b6dca9c262603f6e2a31cfb792b8 (patch)
tree165659456ebcf3f57dd7201b3510ad8f82c71ee3 /Documentation
parentc7e3c93abbc1382923c7f4fe5ba9ea6aa0fa8d0e (diff)
parent4aeba328019ab9fbaad725b923b4c11c3725b24e (diff)
downloadlinux-38e406f600a2b6dca9c262603f6e2a31cfb792b8.tar.bz2
Merge git://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf
Daniel Borkmann says: ==================== pull-request: bpf 2019-06-07 The following pull-request contains BPF updates for your *net* tree. The main changes are: 1) Fix several bugs in riscv64 JIT code emission which forgot to clear high 32-bits for alu32 ops, from Björn and Luke with selftests covering all relevant BPF alu ops from Björn and Jiong. 2) Two fixes for UDP BPF reuseport that avoid calling the program in case of __udp6_lib_err and UDP GRO which broke reuseport_select_sock() assumption that skb->data is pointing to transport header, from Martin. 3) Two fixes for BPF sockmap: a use-after-free from sleep in psock's backlog workqueue, and a missing restore of sk_write_space when psock gets dropped, from Jakub and John. 4) Fix unconnected UDP sendmsg hook API which is insufficient as-is since it breaks standard applications like DNS if reverse NAT is not performed upon receive, from Daniel. 5) Fix an out-of-bounds read in __bpf_skc_lookup which in case of AF_INET6 fails to verify that the length of the tuple is long enough, from Lorenz. 6) Fix libbpf's libbpf__probe_raw_btf to return an fd instead of 0/1 (for {un,}successful probe) as that is expected to be propagated as an fd to load_sk_storage_btf() and thus closing the wrong descriptor otherwise, from Michal. 7) Fix bpftool's JSON output for the case when a lookup fails, from Krzesimir. 8) Minor misc fixes in docs, samples and selftests, from various others. ==================== Signed-off-by: David S. Miller <davem@davemloft.net>
Diffstat (limited to 'Documentation')
-rw-r--r--Documentation/networking/af_xdp.rst8
1 files changed, 4 insertions, 4 deletions
diff --git a/Documentation/networking/af_xdp.rst b/Documentation/networking/af_xdp.rst
index e14d7d40fc75..50bccbf68308 100644
--- a/Documentation/networking/af_xdp.rst
+++ b/Documentation/networking/af_xdp.rst
@@ -316,16 +316,16 @@ A: When a netdev of a physical NIC is initialized, Linux usually
all the traffic, you can force the netdev to only have 1 queue, queue
id 0, and then bind to queue 0. You can use ethtool to do this::
- sudo ethtool -L <interface> combined 1
+ sudo ethtool -L <interface> combined 1
If you want to only see part of the traffic, you can program the
NIC through ethtool to filter out your traffic to a single queue id
that you can bind your XDP socket to. Here is one example in which
UDP traffic to and from port 4242 are sent to queue 2::
- sudo ethtool -N <interface> rx-flow-hash udp4 fn
- sudo ethtool -N <interface> flow-type udp4 src-port 4242 dst-port \
- 4242 action 2
+ sudo ethtool -N <interface> rx-flow-hash udp4 fn
+ sudo ethtool -N <interface> flow-type udp4 src-port 4242 dst-port \
+ 4242 action 2
A number of other ways are possible all up to the capabilitites of
the NIC you have.