selftests: make order checking verbose in msg_zerocopy selftest

We find that when lock debugging is on, notifications may not come in
order. Thus, we have order checking outputs managed by cfg_verbo

selftests: make order checking verbose in msg_zerocopy selftest

We find that when lock debugging is on, notifications may not come in
order. Thus, we have order checking outputs managed by cfg_verbose, to
avoid too many outputs in this case.

Fixes: 07b65c5b31ce ("test: add msg_zerocopy test")
Signed-off-by: Zijian Zhang <[email protected]>
Signed-off-by: Xiaochun Lu <[email protected]>
Reviewed-by: Willem de Bruijn <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Jakub Kicinski <[email protected]>

show more ...

selftests: fix OOM in msg_zerocopy selftest

In selftests/net/msg_zerocopy.c, it has a while loop keeps calling sendmsg
on a socket with MSG_ZEROCOPY flag, and it will recv the notifications
until th

selftests: fix OOM in msg_zerocopy selftest

In selftests/net/msg_zerocopy.c, it has a while loop keeps calling sendmsg
on a socket with MSG_ZEROCOPY flag, and it will recv the notifications
until the socket is not writable. Typically, it will start the receiving
process after around 30+ sendmsgs. However, as the introduction of commit
dfa2f0483360 ("tcp: get rid of sysctl_tcp_adv_win_scale"), the sender is
always writable and does not get any chance to run recv notifications.
The selftest always exits with OUT_OF_MEMORY because the memory used by
opt_skb exceeds the net.core.optmem_max. Meanwhile, it could be set to a
different value to trigger OOM on older kernels too.

Thus, we introduce "cfg_notification_limit" to force sender to receive
notifications after some number of sendmsgs.

Fixes: 07b65c5b31ce ("test: add msg_zerocopy test")
Signed-off-by: Zijian Zhang <[email protected]>
Signed-off-by: Xiaochun Lu <[email protected]>
Reviewed-by: Willem de Bruijn <[email protected]>
Link: https://patch.msgid.link/[email protected]
Signed-off-by: Jakub Kicinski <[email protected]>

show more ...

selftests/net: relax cpu affinity requirement in msg_zerocopy test

The msg_zerocopy test pins the sender and receiver threads to separate
cores to reduce variance between runs.

But it hardcodes the

selftests/net: relax cpu affinity requirement in msg_zerocopy test

The msg_zerocopy test pins the sender and receiver threads to separate
cores to reduce variance between runs.

But it hardcodes the cores and skips core 0, so it fails on machines
with the selected cores offline, or simply fewer cores.

The test mainly gives code coverage in automated runs. The throughput
of zerocopy ('-z') and non-zerocopy runs is logged for manual
inspection.

Continue even when sched_setaffinity fails. Just log to warn anyone
interpreting the data.

Fixes: 07b65c5b31ce ("test: add msg_zerocopy test")
Reported-by: Colin Ian King <[email protected]>
Signed-off-by: Willem de Bruijn <[email protected]>
Acked-by: Colin Ian King <[email protected]>
Signed-off-by: David S. Miller <[email protected]>

show more ...

selftests: extend zerocopy tests to udp

Both msg_zerocopy and udpgso_bench have udp zerocopy variants.
Exercise these as part of the standard kselftest run.

With udp, msg_zerocopy has no control ch

selftests: extend zerocopy tests to udp

Both msg_zerocopy and udpgso_bench have udp zerocopy variants.
Exercise these as part of the standard kselftest run.

With udp, msg_zerocopy has no control channel. Ensure that the
receiver exits after the sender by accounting for the initial
delay in starting them (in msg_zerocopy.sh).

Signed-off-by: Willem de Bruijn <[email protected]>
Signed-off-by: David S. Miller <[email protected]>

show more ...

selftests/net: reap zerocopy completions passed up as ancillary data.

PF_RDS sockets pass up cookies for zerocopy completion as ancillary
data. Update msg_zerocopy to reap this information.

Signed-

selftests/net: reap zerocopy completions passed up as ancillary data.

PF_RDS sockets pass up cookies for zerocopy completion as ancillary
data. Update msg_zerocopy to reap this information.

Signed-off-by: Sowmini Varadhan <[email protected]>
Acked-by: Willem de Bruijn <[email protected]>
Acked-by: Santosh Shilimkar <[email protected]>
Signed-off-by: David S. Miller <[email protected]>

show more ...

selftests/net: revert the zerocopy Rx path for PF_RDS

In preparation for optimized reception of zerocopy completion,
revert the Rx side changes introduced by Commit dfb8434b0a94
("selftests/net: add

selftests/net: revert the zerocopy Rx path for PF_RDS

In preparation for optimized reception of zerocopy completion,
revert the Rx side changes introduced by Commit dfb8434b0a94
("selftests/net: add zerocopy support for PF_RDS test case")

Signed-off-by: Sowmini Varadhan <[email protected]>
Acked-by: Willem de Bruijn <[email protected]>
Acked-by: Santosh Shilimkar <[email protected]>
Signed-off-by: David S. Miller <[email protected]>

show more ...

selftests/net: add zerocopy support for PF_RDS test case

Send a cookie with sendmsg() on PF_RDS sockets, and process the
returned batched cookies in do_recv_completion()

Signed-off-by: Sowmini Vara

selftests/net: add zerocopy support for PF_RDS test case

Send a cookie with sendmsg() on PF_RDS sockets, and process the
returned batched cookies in do_recv_completion()

Signed-off-by: Sowmini Varadhan <[email protected]>
Acked-by: Willem de Bruijn <[email protected]>
Signed-off-by: David S. Miller <[email protected]>

show more ...

selftests/net: add support for PF_RDS sockets

Add support for basic PF_RDS client-server testing in msg_zerocopy

Signed-off-by: Sowmini Varadhan <[email protected]>
Acked-by: Willem de Br

selftests/net: add support for PF_RDS sockets

Add support for basic PF_RDS client-server testing in msg_zerocopy

Signed-off-by: Sowmini Varadhan <[email protected]>
Acked-by: Willem de Bruijn <[email protected]>
Signed-off-by: David S. Miller <[email protected]>

show more ...

selftests/net: fix bugs in address and port initialization

Address/port initialization should work correctly regardless
of the order in which command line arguments are supplied,
E.g, cfg_port shoul

selftests/net: fix bugs in address and port initialization

Address/port initialization should work correctly regardless
of the order in which command line arguments are supplied,
E.g, cfg_port should be used to connect to the remote host
even if it is processed after -D, src/dst address initialization
should not require that [-4|-6] be specified before
the -S or -D args, receiver should be able to bind to *.<cfg_port>

Achieve this by making sure that the address/port structures
are initialized after all command line options are parsed.

Store cfg_port in host-byte order, and use htons()
to set up the sin_port/sin6_port before bind/connect,
so that the network system calls get the correct values
in network-byte order.

Signed-off-by: Sowmini Varadhan <[email protected]>
Acked-by: Willem de Bruijn <[email protected]>
Signed-off-by: David S. Miller <[email protected]>

show more ...

selftests/net: msg_zerocopy enable build with older kernel headers

Explicitly define SO_EE_ORIGIN_ZEROCOPY.
This makes the test program build with older kernel headers,
e.g. from Debian 9.

Signed-o

selftests/net: msg_zerocopy enable build with older kernel headers

Explicitly define SO_EE_ORIGIN_ZEROCOPY.
This makes the test program build with older kernel headers,
e.g. from Debian 9.

Signed-off-by: Thomas Meyer <[email protected]>
Signed-off-by: Shuah Khan <[email protected]>

show more ...

selftests: correct define in msg_zerocopy.c

The msg_zerocopy test defines SO_ZEROCOPY if necessary, but its value
is inconsistent with the one in asm-generic.h. Correct that.

Also convert one error

selftests: correct define in msg_zerocopy.c

The msg_zerocopy test defines SO_ZEROCOPY if necessary, but its value
is inconsistent with the one in asm-generic.h. Correct that.

Also convert one error to a warning. When the test is complete, report
throughput and close cleanly even if the process did not wait for all
completions.

Reported-by: Dan Melnic <[email protected]>
Signed-off-by: Willem de Bruijn <[email protected]>
Acked-by: Alexei Starovoitov <[email protected]>
Signed-off-by: David S. Miller <[email protected]>

show more ...

test: add msg_zerocopy test

Introduce regression test for msg_zerocopy feature. Send traffic from
one process to another with and without zerocopy.

Evaluate tcp, udp, raw and packet sockets, includ

test: add msg_zerocopy test

Introduce regression test for msg_zerocopy feature. Send traffic from
one process to another with and without zerocopy.

Evaluate tcp, udp, raw and packet sockets, including variants
- udp: corking and corking with mixed copy/zerocopy calls
- raw: with and without hdrincl
- packet: at both raw and dgram level

Test on both ipv4 and ipv6, optionally with ethtool changes to
disable scatter-gather, tx checksum or tso offload. All of these
can affect zerocopy behavior.

The regression test can be run on a single machine if over a veth
pair. Then skb_orphan_frags_rx must be modified to be identical to
skb_orphan_frags to allow forwarding zerocopy locally.

The msg_zerocopy.sh script will setup the veth pair in network
namespaces and run all tests.

Signed-off-by: Willem de Bruijn <[email protected]>
Signed-off-by: David S. Miller <[email protected]>

show more ...