linux.git/samples/bpf/xdpsock_user.c, branch master

bpf, samples: Remove AF_XDP samples

2022-07-05T09:58:44+00:00

Remove the AF_XDP samples from samples/bpf/ as they are dependent on
the AF_XDP support in libbpf. This support has now been removed in the
1.0 release, so these samples cannot be compiled anymore. Please start
to use libxdp instead. It is backwards compatible with the AF_XDP
support that was offered in libbpf. New samples can be found in the
various xdp-project repositories connected to libxdp and by googling.

Signed-off-by: Magnus Karlsson 
Signed-off-by: Daniel Borkmann 
Acked-by: Toke Høiland-Jørgensen 
Acked-by: Maciej Fijalkowski 
Acked-by: Jesper Dangaard Brouer 
Link: https://lore.kernel.org/bpf/20220630093717.8664-1-magnus.karlsson@gmail.com

samples/bpf: Use libbpf 1.0 API mode instead of RLIMIT_MEMLOCK

2022-04-11T03:17:15+00:00

We have switched to memcg-based memory accouting and thus the rlimit is
not needed any more. LIBBPF_STRICT_AUTO_RLIMIT_MEMLOCK was introduced in
libbpf for backward compatibility, so we can use it instead now.

This patch also removes the useless header sys/resource.h from many files
in samples/bpf.

Signed-off-by: Yafang Shao 
Signed-off-by: Andrii Nakryiko 
Link: https://lore.kernel.org/bpf/20220409125958.92629-2-laoar.shao@gmail.com

samples/bpf, xdpsock: Fix race when running for fix duration of time

2022-03-15T15:53:37+00:00

When running xdpsock for a fix duration of time before terminating
using --duration=, there is a race condition that may cause xdpsock
to terminate immediately.

When running for a fixed duration of time the check to determine when to
terminate execution is in is_benchmark_done() and is being executed in
the context of the poller thread,

    if (opt_duration > 0) {
            unsigned long dt = (get_nsecs() - start_time);

            if (dt >= opt_duration)
                    benchmark_done = true;
    }

However start_time is only set after the poller thread have been
created. This leaves a small window when the poller thread is starting
and calls is_benchmark_done() for the first time that start_time is not
yet set. In that case start_time have its initial value of 0 and the
duration check fails as it do not correlate correctly for the
applications start time and immediately sets benchmark_done which in
turn terminates the xdpsock application.

Fix this by setting start_time before creating the poller thread.

Fixes: d3f11b018f6c ("samples/bpf: xdpsock: Add duration option to specify how long to run")
Signed-off-by: Niklas Söderlund 
Signed-off-by: Simon Horman 
Signed-off-by: Daniel Borkmann 
Link: https://lore.kernel.org/bpf/20220315102948.466436-1-niklas.soderlund@corigine.com

samples/bpf: adapt samples/bpf to bpf_xdp_xxx() APIs

2022-01-21T05:22:02+00:00

Use new bpf_xdp_*() APIs across all XDP-related BPF samples.

Signed-off-by: Andrii Nakryiko 
Link: https://lore.kernel.org/r/20220120061422.2710637-5-andrii@kernel.org
Signed-off-by: Alexei Starovoitov

samples/bpf: xdpsock: Add timestamp for Tx-only operation

2022-01-06T01:53:24+00:00

It may be useful to add timestamp for Tx packets for continuous or cyclic
transmit operation. The timestamp and sequence ID of a Tx packet are
stored according to pktgen header format. To enable per-packet timestamp,
use -y|--tstamp option. If timestamp is off, pktgen header is not
included in the UDP payload. This means receiving side can use the magic
number for pktgen for differentiation.

The implementation supports both VLAN tagged and untagged option. By
default, the minimum packet size is set at 64B. However, if VLAN tagged
is on (-V), the minimum packet size is increased to 66B just so to fit
the pktgen_hdr size.

Added hex_dump() into the code path just for future cross-checking.
As before, simply change to "#define DEBUG_HEXDUMP 1" to inspect the
accuracy of TX packet.

Signed-off-by: Ong Boon Leong 
Signed-off-by: Alexei Starovoitov 
Link: https://lore.kernel.org/bpf/20211230035447.523177-8-boon.leong.ong@intel.com

samples/bpf: xdpsock: Add time-out for cleaning Tx

2022-01-06T01:53:24+00:00

When user sets tx-pkt-count and in case where there are invalid Tx frame,
the complete_tx_only_all() process polls indefinitely. So, this patch
adds a time-out mechanism into the process so that the application
can terminate automatically after it retries 3*polling interval duration.

v1->v2:
 Thanks to Jesper's and Song Liu's suggestion.
 - clean-up git message to remove polling log
 - make the Tx time-out retries configurable with 1s granularity

Signed-off-by: Ong Boon Leong 
Signed-off-by: Alexei Starovoitov 
Link: https://lore.kernel.org/bpf/20211230035447.523177-7-boon.leong.ong@intel.com

samples/bpf: xdpsock: Add sched policy and priority support

2022-01-06T01:53:24+00:00

By default, TX schedule policy is SCHED_OTHER (round-robin time-sharing).
To improve TX cyclic scheduling, we add SCHED_FIFO policy and its priority
by using -W FIFO or --policy=FIFO and -U  or --schpri=.

A) From xdpsock --app-stats, for SCHED_OTHER policy:
   $ xdpsock -i eth0 -t -N -z -T 1000 -b 16 -C 100000 -a

                      period     min        ave        max        cycle
   Cyclic TX          1000000    53507      75334      712642     6250

B) For SCHED_FIFO policy and schpri=50:
   $ xdpsock -i eth0 -t -N -z -T 1000 -b 16 -C 100000 -a -W FIFO -U 50

                      period     min        ave        max        cycle
   Cyclic TX          1000000    3699       24859      54397      6250

Signed-off-by: Ong Boon Leong 
Signed-off-by: Alexei Starovoitov 
Link: https://lore.kernel.org/bpf/20211230035447.523177-6-boon.leong.ong@intel.com

samples/bpf: xdpsock: Add cyclic TX operation capability

2022-01-06T01:53:24+00:00

Tx cycle time is in micro-seconds unit. By combining the batch size (-b M)
and Tx cycle time (-T|--tx-cycle N), xdpsock now can transmit batch-size of
packets every N-us periodically. Cyclic TX operation is not applicable if
--poll mode is used.

To transmit 16 packets every 1ms cycle time for total of 100000 packets
silently:
 $ xdpsock -i eth0 -T -N -z -T 1000 -b 16 -C 100000

To print cyclic TX schedule variance stats, use --app-stats|-a:
 $ xdpsock -i eth0 -T -N -z -T 1000 -b 16 -C 100000 -a

 sock0@eth0:0 txonly xdp-drv
                   pps            pkts           0.00
rx                 0              0
tx                 0              100000

                   calls/s        count
rx empty polls     0              0
fill fail polls    0              0
copy tx sendtos    0              0
tx wakeup sendtos  0              6254
opt polls          0              0

                   period     min        ave        max        cycle
Cyclic TX          1000000    53507      75334      712642     6250

Signed-off-by: Ong Boon Leong 
Signed-off-by: Alexei Starovoitov 
Link: https://lore.kernel.org/bpf/20211230035447.523177-5-boon.leong.ong@intel.com

samples/bpf: xdpsock: Add clockid selection support

2022-01-06T01:53:24+00:00

User specifies the clock selection by using -w CLOCK or --clock=CLOCK
where CLOCK=[REALTIME, TAI, BOOTTIME, MONOTONIC].

The default CLOCK selection is MONOTONIC.

The implementation of clock selection parsing is borrowed from
iproute2/tc/q_taprio.c

Signed-off-by: Ong Boon Leong 
Signed-off-by: Alexei Starovoitov 
Link: https://lore.kernel.org/bpf/20211230035447.523177-4-boon.leong.ong@intel.com

samples/bpf: xdpsock: Add Dest and Src MAC setting for Tx-only operation

2022-01-06T01:53:24+00:00

To set Dest MAC address (-G|--tx-dmac) only:
 $ xdpsock -i eth0 -t -N -z -G aa:bb:cc:dd:ee:ff

To set Source MAC address (-H|--tx-smac) only:
 $ xdpsock -i eth0 -t -N -z -H 11:22:33:44:55:66

To set both Dest and Source MAC address:
 $ xdpsock -i eth0 -t -N -z -G aa:bb:cc:dd:ee:ff \
   -H 11:22:33:44:55:66

The default Dest and Source MAC address remain the same as before.

Signed-off-by: Ong Boon Leong 
Signed-off-by: Alexei Starovoitov 
Acked-by: Song Liu 
Acked-by: Jesper Dangaard Brouer 
Link: https://lore.kernel.org/bpf/20211230035447.523177-3-boon.leong.ong@intel.com