[PATCH v1 net-next 0/2] selftest: af_unix: Misc updates.

[PATCH v1 net-next 0/2] selftest: af_unix: Misc updates.

[Kuniyuki Iwashima]

Patch 1 add .gitignore under tools/testing/selftests/net/af_unix/.

Patch 2 make so_peek_off.c less flaky.


Kuniyuki Iwashima (2):
  selftest: af_unix: Create its own .gitignore.
  selftest: af_unix: Extend recv() timeout in so_peek_off.c.

 tools/testing/selftests/net/.gitignore            | 8 --------
 tools/testing/selftests/net/af_unix/.gitignore    | 8 ++++++++
 tools/testing/selftests/net/af_unix/so_peek_off.c | 4 ++--
 3 files changed, 10 insertions(+), 10 deletions(-)
 create mode 100644 tools/testing/selftests/net/af_unix/.gitignore

-- 
2.52.0.460.gd25c4c69ec-goog

[PATCH v1 net-next 1/2] selftest: af_unix: Create its own .gitignore.

[Kuniyuki Iwashima]

Somehow AF_UNIX tests have reused ../.gitignore,
but now NIPA warns about it.

Let's create .gitignore under af_unix/.

Signed-off-by: Kuniyuki Iwashima <kuniyu@google.com>
---
 tools/testing/selftests/net/.gitignore         | 8 --------
 tools/testing/selftests/net/af_unix/.gitignore | 8 ++++++++
 2 files changed, 8 insertions(+), 8 deletions(-)
 create mode 100644 tools/testing/selftests/net/af_unix/.gitignore

diff --git a/tools/testing/selftests/net/.gitignore b/tools/testing/selftests/net/.gitignore
index 8779ad303748..df790867d479 100644
--- a/tools/testing/selftests/net/.gitignore
+++ b/tools/testing/selftests/net/.gitignore
@@ -4,7 +4,6 @@ bind_timewait
 bind_wildcard
 busy_poller
 cmsg_sender
-diag_uid
 epoll_busy_poll
 fin_ack_lat
 hwtstamp_config
@@ -17,7 +16,6 @@ ipv6_flowlabel
 ipv6_flowlabel_mgr
 ipv6_fragmentation
 log.txt
-msg_oob
 msg_zerocopy
 netlink-dumps
 nettest
@@ -35,9 +33,6 @@ reuseport_dualstack
 reuseport_priority
 rxtimestamp
 sctp_hello
-scm_inq
-scm_pidfd
-scm_rights
 sk_bind_sendto_listen
 sk_connect_zero_addr
 sk_so_peek_off
@@ -45,7 +40,6 @@ skf_net_off
 socket
 so_incoming_cpu
 so_netns_cookie
-so_peek_off
 so_txtime
 so_rcv_listener
 stress_reuseport_listen
@@ -64,5 +58,3 @@ txtimestamp
 udpgso
 udpgso_bench_rx
 udpgso_bench_tx
-unix_connect
-unix_connreset
diff --git a/tools/testing/selftests/net/af_unix/.gitignore b/tools/testing/selftests/net/af_unix/.gitignore
new file mode 100644
index 000000000000..694bcb11695b
--- /dev/null
+++ b/tools/testing/selftests/net/af_unix/.gitignore
@@ -0,0 +1,8 @@
+diag_uid
+msg_oob
+scm_inq
+scm_pidfd
+scm_rights
+so_peek_off
+unix_connect
+unix_connreset
\ No newline at end of file
-- 
2.52.0.460.gd25c4c69ec-goog

Re: [PATCH v1 net-next 1/2] selftest: af_unix: Create its own .gitignore.

[Jakub Kicinski]

On Mon, 24 Nov 2025 19:43:33 +0000 Kuniyuki Iwashima wrote:
> Somehow AF_UNIX tests have reused ../.gitignore,
> but now NIPA warns about it.
> 
> Let's create .gitignore under af_unix/.

Thanks for following up!

NIPA says it doesn't apply:

error: patch failed: tools/testing/selftests/net/.gitignore:35
error: tools/testing/selftests/net/.gitignore: patch does not apply
hint: Use 'git am --show-current-patch=diff' to see the failed patch
hint: When you have resolved this problem, run "git am --continue".
hint: If you prefer to skip this patch, run "git am --skip" instead.
hint: To restore the original branch and stop patching, run "git am --abort".
hint: Disable this message with "git config set advice.mergeConflict false"

Does it have a dependency in net or on the list?

> new file mode 100644
> index 000000000000..694bcb11695b
> --- /dev/null
> +++ b/tools/testing/selftests/net/af_unix/.gitignore
> @@ -0,0 +1,8 @@
> +diag_uid
> +msg_oob
> +scm_inq
> +scm_pidfd
> +scm_rights
> +so_peek_off
> +unix_connect
> +unix_connreset
> \ No newline at end of file

nit: missing new line
-- 
pw-bot: cr

Re: [PATCH v1 net-next 1/2] selftest: af_unix: Create its own .gitignore.

[Kuniyuki Iwashima]

On Mon, Nov 24, 2025 at 11:51 AM Jakub Kicinski <kuba@kernel.org> wrote:
>
> On Mon, 24 Nov 2025 19:43:33 +0000 Kuniyuki Iwashima wrote:
> > Somehow AF_UNIX tests have reused ../.gitignore,
> > but now NIPA warns about it.
> >
> > Let's create .gitignore under af_unix/.
>
> Thanks for following up!
>
> NIPA says it doesn't apply:
>
> error: patch failed: tools/testing/selftests/net/.gitignore:35
> error: tools/testing/selftests/net/.gitignore: patch does not apply
> hint: Use 'git am --show-current-patch=diff' to see the failed patch
> hint: When you have resolved this problem, run "git am --continue".
> hint: If you prefer to skip this patch, run "git am --skip" instead.
> hint: To restore the original branch and stop patching, run "git am --abort".
> hint: Disable this message with "git config set advice.mergeConflict false"
>
> Does it have a dependency in net or on the list?

Oh sorry, I forgot I had some downstream tests.
Will rebase on clean net-next.

>
> > new file mode 100644
> > index 000000000000..694bcb11695b
> > --- /dev/null
> > +++ b/tools/testing/selftests/net/af_unix/.gitignore
> > @@ -0,0 +1,8 @@
> > +diag_uid
> > +msg_oob
> > +scm_inq
> > +scm_pidfd
> > +scm_rights
> > +so_peek_off
> > +unix_connect
> > +unix_connreset
> > \ No newline at end of file
>
> nit: missing new line

Will add in v2.

Thanks!

Re: [PATCH v1 net-next 1/2] selftest: af_unix: Create its own .gitignore.

[Jakub Kicinski]

On Mon, 24 Nov 2025 12:07:28 -0800 Kuniyuki Iwashima wrote:
> > nit: missing new line  
> 
> Will add in v2.

Feel free to skip the 24h wait. I'd like to get the netdev foundation CI
to a clean state so I can shut off the AWS one completely.

Re: [PATCH v1 net-next 1/2] selftest: af_unix: Create its own .gitignore.

[Kuniyuki Iwashima]

On Mon, Nov 24, 2025 at 1:12 PM Jakub Kicinski <kuba@kernel.org> wrote:
>
> On Mon, 24 Nov 2025 12:07:28 -0800 Kuniyuki Iwashima wrote:
> > > nit: missing new line
> >
> > Will add in v2.
>
> Feel free to skip the 24h wait. I'd like to get the netdev foundation CI
> to a clean state so I can shut off the AWS one completely.

Thanks, I'll respin shortly.

[PATCH v1 net-next 2/2] selftest: af_unix: Extend recv() timeout in so_peek_off.c.

[Kuniyuki Iwashima]

so_peek_off.c is reported to be flaky on NIPA:

  # # so_peek_off.c:149:two_chunks_overlap_blocking:Expected -1 (-1) != bytes (-1)
  # # two_chunks_overlap_blocking: Test terminated by assertion
  # #          FAIL  so_peek_off.stream.two_chunks_overlap_blocking

The test fork()s a child process to send() data after 1ms to
wake up the parent process being blocked (up to 3ms) on recv().

But, from the log, the parent woke up after 3ms timeout, so it
could be too short when the host is overloaded.

Let's extend it to 5s.

Reported-by: Jakub Kicinski <kuba@kernel.org>
Closes: https://lore.kernel.org/netdev/20251124070722.1e828c53@kernel.org/
Signed-off-by: Kuniyuki Iwashima <kuniyu@google.com>
---
 tools/testing/selftests/net/af_unix/so_peek_off.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/tools/testing/selftests/net/af_unix/so_peek_off.c b/tools/testing/selftests/net/af_unix/so_peek_off.c
index 1a77728128e5..86e7b0fb522d 100644
--- a/tools/testing/selftests/net/af_unix/so_peek_off.c
+++ b/tools/testing/selftests/net/af_unix/so_peek_off.c
@@ -36,8 +36,8 @@ FIXTURE_VARIANT_ADD(so_peek_off, seqpacket)
 FIXTURE_SETUP(so_peek_off)
 {
 	struct timeval timeout = {
-		.tv_sec = 0,
-		.tv_usec = 3000,
+		.tv_sec = 5,
+		.tv_usec = 0,
 	};
 	int ret;
 
-- 
2.52.0.460.gd25c4c69ec-goog