From: Jakub Kicinski <kuba@kernel.org>
To: davem@davemloft.net
Cc: netdev@vger.kernel.org, edumazet@google.com, pabeni@redhat.com,
Jakub Kicinski <kuba@kernel.org>,
shuah@kernel.org, linux-kselftest@vger.kernel.org
Subject: [PATCH net-next 2/2] selftests: net: increase the delay for relative cmsg_time.sh test
Date: Thu, 9 May 2024 17:57:05 -0700 [thread overview]
Message-ID: <20240510005705.43069-2-kuba@kernel.org> (raw)
In-Reply-To: <20240510005705.43069-1-kuba@kernel.org>
Slow machines can delay scheduling of the packets for milliseconds.
Increase the delay to 8ms if KSFT_MACHINE_SLOW. Try to limit the
variability by moving setsockopts earlier (before we read time).
This fixes the "TXTIME rel" failures on debug kernels, like:
Case ICMPv4 - TXTIME rel returned '', expected 'OK'
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
CC: shuah@kernel.org
CC: linux-kselftest@vger.kernel.org
---
tools/testing/selftests/net/cmsg_sender.c | 32 +++++++++++++----------
tools/testing/selftests/net/cmsg_time.sh | 7 +++--
2 files changed, 23 insertions(+), 16 deletions(-)
diff --git a/tools/testing/selftests/net/cmsg_sender.c b/tools/testing/selftests/net/cmsg_sender.c
index f25268504937..b2df05ef71cb 100644
--- a/tools/testing/selftests/net/cmsg_sender.c
+++ b/tools/testing/selftests/net/cmsg_sender.c
@@ -260,15 +260,8 @@ cs_write_cmsg(int fd, struct msghdr *msg, char *cbuf, size_t cbuf_sz)
SOL_IPV6, IPV6_HOPLIMIT, &opt.v6.hlimit);
if (opt.txtime.ena) {
- struct sock_txtime so_txtime = {
- .clockid = CLOCK_MONOTONIC,
- };
__u64 txtime;
- if (setsockopt(fd, SOL_SOCKET, SO_TXTIME,
- &so_txtime, sizeof(so_txtime)))
- error(ERN_SOCKOPT, errno, "setsockopt TXTIME");
-
txtime = time_start_mono.tv_sec * (1000ULL * 1000 * 1000) +
time_start_mono.tv_nsec +
opt.txtime.delay * 1000;
@@ -284,13 +277,6 @@ cs_write_cmsg(int fd, struct msghdr *msg, char *cbuf, size_t cbuf_sz)
memcpy(CMSG_DATA(cmsg), &txtime, sizeof(txtime));
}
if (opt.ts.ena) {
- __u32 val = SOF_TIMESTAMPING_SOFTWARE |
- SOF_TIMESTAMPING_OPT_TSONLY;
-
- if (setsockopt(fd, SOL_SOCKET, SO_TIMESTAMPING,
- &val, sizeof(val)))
- error(ERN_SOCKOPT, errno, "setsockopt TIMESTAMPING");
-
cmsg = (struct cmsghdr *)(cbuf + cmsg_len);
cmsg_len += CMSG_SPACE(sizeof(__u32));
if (cbuf_sz < cmsg_len)
@@ -426,6 +412,24 @@ static void ca_set_sockopts(int fd)
setsockopt(fd, SOL_SOCKET, SO_PRIORITY,
&opt.sockopt.priority, sizeof(opt.sockopt.priority)))
error(ERN_SOCKOPT, errno, "setsockopt SO_PRIORITY");
+
+ if (opt.txtime.ena) {
+ struct sock_txtime so_txtime = {
+ .clockid = CLOCK_MONOTONIC,
+ };
+
+ if (setsockopt(fd, SOL_SOCKET, SO_TXTIME,
+ &so_txtime, sizeof(so_txtime)))
+ error(ERN_SOCKOPT, errno, "setsockopt TXTIME");
+ }
+ if (opt.ts.ena) {
+ __u32 val = SOF_TIMESTAMPING_SOFTWARE |
+ SOF_TIMESTAMPING_OPT_TSONLY;
+
+ if (setsockopt(fd, SOL_SOCKET, SO_TIMESTAMPING,
+ &val, sizeof(val)))
+ error(ERN_SOCKOPT, errno, "setsockopt TIMESTAMPING");
+ }
}
int main(int argc, char *argv[])
diff --git a/tools/testing/selftests/net/cmsg_time.sh b/tools/testing/selftests/net/cmsg_time.sh
index af85267ad1e3..1d7e756644bc 100755
--- a/tools/testing/selftests/net/cmsg_time.sh
+++ b/tools/testing/selftests/net/cmsg_time.sh
@@ -66,10 +66,13 @@ for i in "-4 $TGT4" "-6 $TGT6"; do
awk '/SND/ { if ($3 > 1000) print "OK"; }')
check_result $? "$ts" "OK" "$prot - TXTIME abs"
- ts=$(ip netns exec $NS ./cmsg_sender -p $p $i 1234 -t -d 1000 |
+ [ "$KSFT_MACHINE_SLOW" = yes ] && delay=8000 || delay=1000
+
+ ts=$(ip netns exec $NS ./cmsg_sender -p $p $i 1234 -t -d $delay |
awk '/SND/ {snd=$3}
/SCHED/ {sch=$3}
- END { if (snd - sch > 500) print "OK"; }')
+ END { if (snd - sch > '$((delay/2))') print "OK";
+ else print snd, "-", sch, "<", '$((delay/2))'; }')
check_result $? "$ts" "OK" "$prot - TXTIME rel"
done
done
--
2.45.0
next prev parent reply other threads:[~2024-05-10 0:58 UTC|newest]
Thread overview: 6+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-05-10 0:57 [PATCH net-next 1/2] selftests: net: fix timestamp not arriving in cmsg_time.sh Jakub Kicinski
2024-05-10 0:57 ` Jakub Kicinski [this message]
2024-05-10 2:25 ` [PATCH net-next 2/2] selftests: net: increase the delay for relative cmsg_time.sh test Willem de Bruijn
2024-05-10 2:49 ` [PATCH net-next 1/2] selftests: net: fix timestamp not arriving in cmsg_time.sh Willem de Bruijn
2024-05-11 1:23 ` Jakub Kicinski
2024-05-11 1:30 ` patchwork-bot+netdevbpf
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20240510005705.43069-2-kuba@kernel.org \
--to=kuba@kernel.org \
--cc=davem@davemloft.net \
--cc=edumazet@google.com \
--cc=linux-kselftest@vger.kernel.org \
--cc=netdev@vger.kernel.org \
--cc=pabeni@redhat.com \
--cc=shuah@kernel.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.