[PATCH net-next 0/3] selftests/xsk: stabilize timeout test behavior

[PATCH net-next 0/3] selftests/xsk: stabilize timeout test behavior

[Tushar Vyavahare]

This series improves AF_XDP selftests by making timeout handling
explicit and fixing sources of non-determinism in xsk timeout tests.

Patch 1 introduces test_spec::poll_tmout and removes implicit
dependence on RX UMEM setup state for timeout behavior.

Patch 2 fixes thread harness sequencing by attaching XDP programs
before worker startup, removing signal-based termination, and using
barrier synchronization only for dual-thread runs.

Patch 3 restores shared_umem after POLL_TXQ_FULL so test-local
configuration does not leak into subsequent cases on shared-netdev
runs.

Together these changes make timeout handling easier to follow and
improve selftest stability, especially on real NIC runs.

Tushar Vyavahare (3):
  selftests/xsk: make poll timeout mode explicit
  selftests/xsk: fix timeout thread harness sequencing
  selftests/xsk: restore shared_umem after POLL_TXQ_FULL

 .../selftests/bpf/prog_tests/test_xsk.c       | 96 +++++++++++--------
 .../selftests/bpf/prog_tests/test_xsk.h       |  2 +
 2 files changed, 56 insertions(+), 42 deletions(-)

-- 
2.43.0

[PATCH net-next 1/3] selftests/xsk: make poll timeout mode explicit

[Tushar Vyavahare]

Stop inferring timeout behavior from RX UMEM initialization state.
That ties timeout semantics to setup internals and obscures intent.

Use test_spec::poll_tmout as the explicit timeout-mode selector in
TX and RX paths.

In RX, treat poll timeout as expected only in timeout mode.
In TX, let send_pkts() own loop completion in non-timeout mode
and use __send_pkts() only for progress and timeout detection.

This makes timeout logic explicit and keeps control flow predictable.

Signed-off-by: Magnus Karlsson <magnus.karlsson@intel.com>
Signed-off-by: Tushar Vyavahare <tushar.vyavahare@intel.com>
---
 .../selftests/bpf/prog_tests/test_xsk.c       | 44 +++++++++----------
 .../selftests/bpf/prog_tests/test_xsk.h       |  1 +
 2 files changed, 21 insertions(+), 24 deletions(-)

diff --git a/tools/testing/selftests/bpf/prog_tests/test_xsk.c b/tools/testing/selftests/bpf/prog_tests/test_xsk.c
index 72875071d4f1..ca47a16ceb1a 100644
--- a/tools/testing/selftests/bpf/prog_tests/test_xsk.c
+++ b/tools/testing/selftests/bpf/prog_tests/test_xsk.c
@@ -65,11 +65,6 @@ static void gen_eth_hdr(struct xsk_socket_info *xsk, struct ethhdr *eth_hdr)
 	eth_hdr->h_proto = htons(ETH_P_LOOPBACK);
 }
 
-static bool is_umem_valid(struct xsk_socket_info *xsk)
-{
-	return !!xsk->umem->umem;
-}
-
 static u32 mode_to_xdp_flags(enum test_mode mode)
 {
 	return (mode == TEST_MODE_SKB) ? XDP_FLAGS_SKB_MODE : XDP_FLAGS_DRV_MODE;
@@ -1010,7 +1005,7 @@ static int __receive_pkts(struct test_spec *test, struct xsk_socket_info *xsk)
 			return TEST_FAILURE;
 
 		if (!ret) {
-			if (!is_umem_valid(test->ifobj_tx->xsk))
+			if (test->poll_tmout)
 				return TEST_PASS;
 
 			ksft_print_msg("ERROR: [%s] Poll timed out\n", __func__);
@@ -1149,7 +1144,7 @@ static int receive_pkts(struct test_spec *test)
 			break;
 
 		res = __receive_pkts(test, xsk);
-		if (!(res == TEST_PASS || res == TEST_CONTINUE))
+		if (res != TEST_CONTINUE)
 			return res;
 
 		ret = gettimeofday(&tv_now, NULL);
@@ -1166,7 +1161,8 @@ static int receive_pkts(struct test_spec *test)
 	return TEST_PASS;
 }
 
-static int __send_pkts(struct ifobject *ifobject, struct xsk_socket_info *xsk, bool timeout)
+static int __send_pkts(struct ifobject *ifobject, struct xsk_socket_info *xsk,
+		       bool test_timeout)
 {
 	u32 i, idx = 0, valid_pkts = 0, valid_frags = 0, buffer_len;
 	struct pkt_stream *pkt_stream = xsk->pkt_stream;
@@ -1178,7 +1174,7 @@ static int __send_pkts(struct ifobject *ifobject, struct xsk_socket_info *xsk, b
 	buffer_len = pkt_get_buffer_len(umem, pkt_stream->max_pkt_len);
 	/* pkts_in_flight might be negative if many invalid packets are sent */
 	if (pkts_in_flight >= (int)((umem_size(umem) - xsk->batch_size * buffer_len) /
-	    buffer_len)) {
+	    buffer_len) && !test_timeout) {
 		ret = kick_tx(xsk);
 		if (ret)
 			return TEST_FAILURE;
@@ -1191,7 +1187,7 @@ static int __send_pkts(struct ifobject *ifobject, struct xsk_socket_info *xsk, b
 	while (xsk_ring_prod__reserve(&xsk->tx, xsk->batch_size, &idx) < xsk->batch_size) {
 		if (use_poll) {
 			ret = poll(&fds, 1, POLL_TMOUT);
-			if (timeout) {
+			if (test_timeout) {
 				if (ret < 0) {
 					ksft_print_msg("ERROR: [%s] Poll error %d\n",
 						       __func__, errno);
@@ -1271,7 +1267,7 @@ static int __send_pkts(struct ifobject *ifobject, struct xsk_socket_info *xsk, b
 	if (use_poll) {
 		ret = poll(&fds, 1, POLL_TMOUT);
 		if (ret <= 0) {
-			if (ret == 0 && timeout)
+			if (ret == 0 && test_timeout)
 				return TEST_PASS;
 
 			ksft_print_msg("ERROR: [%s] Poll error %d\n", __func__, ret);
@@ -1279,14 +1275,14 @@ static int __send_pkts(struct ifobject *ifobject, struct xsk_socket_info *xsk, b
 		}
 	}
 
-	if (!timeout) {
+	if (!test_timeout) {
 		if (complete_pkts(xsk, i))
 			return TEST_FAILURE;
 
 		usleep(10);
-		return TEST_PASS;
 	}
 
+	/* Loop completion is driven by send_pkts() stream progress checks. */
 	return TEST_CONTINUE;
 }
 
@@ -1322,7 +1318,6 @@ bool all_packets_sent(struct test_spec *test, unsigned long *bitmap)
 
 static int send_pkts(struct test_spec *test, struct ifobject *ifobject)
 {
-	bool timeout = !is_umem_valid(test->ifobj_rx->xsk);
 	DECLARE_BITMAP(bitmap, test->nb_sockets);
 	u32 i, ret;
 
@@ -1337,19 +1332,18 @@ static int send_pkts(struct test_spec *test, struct ifobject *ifobject)
 				__set_bit(i, bitmap);
 				continue;
 			}
-			ret = __send_pkts(ifobject, &ifobject->xsk_arr[i], timeout);
-			if (ret == TEST_CONTINUE && !test->fail)
-				continue;
-
-			if ((ret || test->fail) && !timeout)
-				return TEST_FAILURE;
-
-			if (ret == TEST_PASS && timeout)
+			ret = __send_pkts(ifobject, &ifobject->xsk_arr[i], test->poll_tmout);
+			if (ret != TEST_CONTINUE)
 				return ret;
 
-			ret = wait_for_tx_completion(&ifobject->xsk_arr[i]);
-			if (ret)
+			if (test->fail)
 				return TEST_FAILURE;
+
+			if (!test->poll_tmout) {
+				ret = wait_for_tx_completion(&ifobject->xsk_arr[i]);
+				if (ret)
+					return TEST_FAILURE;
+			}
 		}
 	}
 
@@ -2231,6 +2225,7 @@ int testapp_xdp_shared_umem(struct test_spec *test)
 
 int testapp_poll_txq_tmout(struct test_spec *test)
 {
+	test->poll_tmout = true;
 	test->ifobj_tx->use_poll = true;
 	/* create invalid frame by set umem frame_size and pkt length equal to 2048 */
 	test->ifobj_tx->xsk->umem->frame_size = 2048;
@@ -2241,6 +2236,7 @@ int testapp_poll_txq_tmout(struct test_spec *test)
 
 int testapp_poll_rxq_tmout(struct test_spec *test)
 {
+	test->poll_tmout = true;
 	test->ifobj_rx->use_poll = true;
 	return testapp_validate_traffic_single_thread(test, test->ifobj_rx);
 }
diff --git a/tools/testing/selftests/bpf/prog_tests/test_xsk.h b/tools/testing/selftests/bpf/prog_tests/test_xsk.h
index 4313d0d87235..20eaaa254998 100644
--- a/tools/testing/selftests/bpf/prog_tests/test_xsk.h
+++ b/tools/testing/selftests/bpf/prog_tests/test_xsk.h
@@ -207,6 +207,7 @@ struct test_spec {
 	bool set_ring;
 	bool adjust_tail;
 	bool adjust_tail_support;
+	bool poll_tmout;
 	enum test_mode mode;
 	char name[MAX_TEST_NAME_SIZE];
 };
-- 
2.43.0

[PATCH net-next 2/3] selftests/xsk: fix timeout thread harness sequencing

[Tushar Vyavahare]

Prevent workers from running before XDP program attachment completes.
The previous ordering allowed races between worker startup and setup.

Attach XDP programs before entering traffic validation.

Remove SIGUSR1-based worker termination and use pthread_join() for
thread shutdown so blocking syscalls are not interrupted.

Use barriers only for dual-thread runs so participants match and
teardown ordering stays deterministic.

This removes setup/startup races and stabilizes harness sequencing.

Signed-off-by: Tushar Vyavahare <tushar.vyavahare@intel.com>
---
 .../selftests/bpf/prog_tests/test_xsk.c       | 33 ++++++++++---------
 .../selftests/bpf/prog_tests/test_xsk.h       |  1 +
 2 files changed, 18 insertions(+), 16 deletions(-)

diff --git a/tools/testing/selftests/bpf/prog_tests/test_xsk.c b/tools/testing/selftests/bpf/prog_tests/test_xsk.c
index ca47a16ceb1a..d4702d2aac5e 100644
--- a/tools/testing/selftests/bpf/prog_tests/test_xsk.c
+++ b/tools/testing/selftests/bpf/prog_tests/test_xsk.c
@@ -7,7 +7,6 @@
 #include <linux/netdev.h>
 #include <poll.h>
 #include <pthread.h>
-#include <signal.h>
 #include <string.h>
 #include <sys/mman.h>
 #include <sys/socket.h>
@@ -1671,7 +1670,8 @@ void *worker_testapp_validate_rx(void *arg)
 				       strerror(-err));
 	}
 
-	pthread_barrier_wait(&barr);
+	if (test->use_barrier)
+		pthread_barrier_wait(&barr);
 
 	/* We leave only now in case of error to avoid getting stuck in the barrier */
 	if (err) {
@@ -1710,11 +1710,6 @@ static void testapp_clean_xsk_umem(struct ifobject *ifobj)
 	munmap(umem->buffer, umem->mmap_size);
 }
 
-static void handler(int signum)
-{
-	pthread_exit(NULL);
-}
-
 static bool xdp_prog_changed_rx(struct test_spec *test)
 {
 	struct ifobject *ifobj = test->ifobj_rx;
@@ -1819,9 +1814,18 @@ static int __testapp_validate_traffic(struct test_spec *test, struct ifobject *i
 		return TEST_FAILURE;
 	}
 
-	if (ifobj2) {
+	err = xsk_attach_xdp_progs(test, ifobj1, ifobj2);
+	if (err) {
+		ksft_print_msg("Error: failed to attach XDP programs: %d (%s)\n",
+			       err, strerror(-err));
+		return TEST_FAILURE;
+	}
+	test->use_barrier = !!ifobj2;
+
+	if (test->use_barrier) {
 		if (pthread_barrier_init(&barr, NULL, 2))
 			return TEST_FAILURE;
+
 		pkt_stream_reset(ifobj2->xsk->pkt_stream);
 	}
 
@@ -1829,27 +1833,26 @@ static int __testapp_validate_traffic(struct test_spec *test, struct ifobject *i
 	pkt_stream_reset(ifobj1->xsk->pkt_stream);
 	pkts_in_flight = 0;
 
-	signal(SIGUSR1, handler);
 	/*Spawn RX thread */
 	pthread_create(&t0, NULL, ifobj1->func_ptr, test);
 
-	if (ifobj2) {
+	if (test->use_barrier) {
 		pthread_barrier_wait(&barr);
 		if (pthread_barrier_destroy(&barr)) {
-			pthread_kill(t0, SIGUSR1);
+			test->use_barrier = false;
+			pthread_join(t0, NULL);
 			clean_sockets(test, ifobj1);
 			clean_umem(test, ifobj1, NULL);
 			return TEST_FAILURE;
 		}
+	}
 
+	if (ifobj2) {
 		/*Spawn TX thread */
 		pthread_create(&t1, NULL, ifobj2->func_ptr, test);
-
 		pthread_join(t1, NULL);
 	}
 
-	if (!ifobj2)
-		pthread_kill(t0, SIGUSR1);
 	pthread_join(t0, NULL);
 
 	if (test->total_steps == test->current_step || test->fail) {
@@ -1887,8 +1890,6 @@ static int testapp_validate_traffic(struct test_spec *test)
 		}
 	}
 
-	if (xsk_attach_xdp_progs(test, ifobj_rx, ifobj_tx))
-		return TEST_FAILURE;
 	return __testapp_validate_traffic(test, ifobj_rx, ifobj_tx);
 }
 
diff --git a/tools/testing/selftests/bpf/prog_tests/test_xsk.h b/tools/testing/selftests/bpf/prog_tests/test_xsk.h
index 20eaaa254998..03753ddc5dcd 100644
--- a/tools/testing/selftests/bpf/prog_tests/test_xsk.h
+++ b/tools/testing/selftests/bpf/prog_tests/test_xsk.h
@@ -208,6 +208,7 @@ struct test_spec {
 	bool adjust_tail;
 	bool adjust_tail_support;
 	bool poll_tmout;
+	bool use_barrier;
 	enum test_mode mode;
 	char name[MAX_TEST_NAME_SIZE];
 };
-- 
2.43.0

[PATCH net-next 3/3] selftests/xsk: restore shared_umem after POLL_TXQ_FULL

[Tushar Vyavahare]

POLL_TXQ_FULL temporarily disables shared_umem on TX to exercise the
TX timeout path in isolation.

With shared_umem enabled, TX setup expects RX UMEM to be initialized
first and fails with: "RX UMEM is not initialized before shared-UMEM TX
setup".

Save and restore shared_umem around POLL_TXQ_FULL execution, and restore
it on both success and pkt_stream_replace() failure paths.

Also add an in-code comment explaining why shared_umem is temporarily
disabled in this test.

This keeps timeout setup local and prevents cross-test state leakage.

Signed-off-by: Tushar Vyavahare <tushar.vyavahare@intel.com>
---
 .../selftests/bpf/prog_tests/test_xsk.c       | 19 +++++++++++++++++--
 1 file changed, 17 insertions(+), 2 deletions(-)

diff --git a/tools/testing/selftests/bpf/prog_tests/test_xsk.c b/tools/testing/selftests/bpf/prog_tests/test_xsk.c
index d4702d2aac5e..6eb9096d084c 100644
--- a/tools/testing/selftests/bpf/prog_tests/test_xsk.c
+++ b/tools/testing/selftests/bpf/prog_tests/test_xsk.c
@@ -2226,13 +2226,28 @@ int testapp_xdp_shared_umem(struct test_spec *test)
 
 int testapp_poll_txq_tmout(struct test_spec *test)
 {
+	bool shared_umem = test->ifobj_tx->shared_umem;
+	int ret;
+
 	test->poll_tmout = true;
+	/*
+	 * POLL_TXQ_FULL exercises TX timeout setup in isolation.
+	 * Keep TX out of shared-UMEM mode here so TX setup does not require
+	 * RX UMEM to be initialized first.
+	 */
+	test->ifobj_tx->shared_umem = false;
 	test->ifobj_tx->use_poll = true;
 	/* create invalid frame by set umem frame_size and pkt length equal to 2048 */
 	test->ifobj_tx->xsk->umem->frame_size = 2048;
-	if (pkt_stream_replace(test, 2 * DEFAULT_PKT_CNT, 2048))
+	if (pkt_stream_replace(test, 2 * DEFAULT_PKT_CNT, 2048)) {
+		test->ifobj_tx->shared_umem = shared_umem;
 		return TEST_FAILURE;
-	return testapp_validate_traffic_single_thread(test, test->ifobj_tx);
+	}
+
+	ret = testapp_validate_traffic_single_thread(test, test->ifobj_tx);
+	test->ifobj_tx->shared_umem = shared_umem;
+
+	return ret;
 }
 
 int testapp_poll_rxq_tmout(struct test_spec *test)
-- 
2.43.0

Re: [PATCH net-next 0/3] selftests/xsk: stabilize timeout test behavior

[Jason Xing]

Hi Tushar,

On Tue, Jun 16, 2026 at 11:50 PM Tushar Vyavahare
<tushar.vyavahare@intel.com> wrote:
>
> This series improves AF_XDP selftests by making timeout handling
> explicit and fixing sources of non-determinism in xsk timeout tests.
>
> Patch 1 introduces test_spec::poll_tmout and removes implicit
> dependence on RX UMEM setup state for timeout behavior.
>
> Patch 2 fixes thread harness sequencing by attaching XDP programs
> before worker startup, removing signal-based termination, and using
> barrier synchronization only for dual-thread runs.
>
> Patch 3 restores shared_umem after POLL_TXQ_FULL so test-local
> configuration does not leak into subsequent cases on shared-netdev
> runs.
>
> Together these changes make timeout handling easier to follow and
> improve selftest stability, especially on real NIC runs.

net-next is closed, but in the meantime I'll review the series ASAP.

BTW, another thing about selftests I had in my mind is that are you
planning to work on this [1]?

[1]: https://lore.kernel.org/all/20260520004244.55663-1-kerneljasonxing@gmail.com/

Thanks,
Jason

>
> Tushar Vyavahare (3):
>   selftests/xsk: make poll timeout mode explicit
>   selftests/xsk: fix timeout thread harness sequencing
>   selftests/xsk: restore shared_umem after POLL_TXQ_FULL
>
>  .../selftests/bpf/prog_tests/test_xsk.c       | 96 +++++++++++--------
>  .../selftests/bpf/prog_tests/test_xsk.h       |  2 +
>  2 files changed, 56 insertions(+), 42 deletions(-)
>
> --
> 2.43.0
>
>

Re: [PATCH net-next 0/3] selftests/xsk: stabilize timeout test behavior

[Maciej Fijalkowski]

On Wed, Jun 17, 2026 at 07:39:06AM +0800, Jason Xing wrote:
> Hi Tushar,
> 
> On Tue, Jun 16, 2026 at 11:50 PM Tushar Vyavahare
> <tushar.vyavahare@intel.com> wrote:
> >
> > This series improves AF_XDP selftests by making timeout handling
> > explicit and fixing sources of non-determinism in xsk timeout tests.
> >
> > Patch 1 introduces test_spec::poll_tmout and removes implicit
> > dependence on RX UMEM setup state for timeout behavior.
> >
> > Patch 2 fixes thread harness sequencing by attaching XDP programs
> > before worker startup, removing signal-based termination, and using
> > barrier synchronization only for dual-thread runs.
> >
> > Patch 3 restores shared_umem after POLL_TXQ_FULL so test-local
> > configuration does not leak into subsequent cases on shared-netdev
> > runs.
> >
> > Together these changes make timeout handling easier to follow and
> > improve selftest stability, especially on real NIC runs.
> 
> net-next is closed, but in the meantime I'll review the series ASAP.
> 
> BTW, another thing about selftests I had in my mind is that are you
> planning to work on this [1]?

This one is on me. I took your changes Jason and aligned ZC batching side
to this behavior, followed by xskxceiver adjustment. I am planning to send
this today EOD, however let's see how badly internal Sashiko will kick my
ass.

> 
> [1]: https://lore.kernel.org/all/20260520004244.55663-1-kerneljasonxing@gmail.com/
> 
> Thanks,
> Jason
> 
> >
> > Tushar Vyavahare (3):
> >   selftests/xsk: make poll timeout mode explicit
> >   selftests/xsk: fix timeout thread harness sequencing
> >   selftests/xsk: restore shared_umem after POLL_TXQ_FULL
> >
> >  .../selftests/bpf/prog_tests/test_xsk.c       | 96 +++++++++++--------
> >  .../selftests/bpf/prog_tests/test_xsk.h       |  2 +
> >  2 files changed, 56 insertions(+), 42 deletions(-)
> >
> > --
> > 2.43.0
> >
> >
> 

Re: [PATCH net-next 0/3] selftests/xsk: stabilize timeout test behavior

[Jason Xing]

On Wed, Jun 17, 2026 at 5:43 PM Maciej Fijalkowski
<maciej.fijalkowski@intel.com> wrote:
>
> On Wed, Jun 17, 2026 at 07:39:06AM +0800, Jason Xing wrote:
> > Hi Tushar,
> >
> > On Tue, Jun 16, 2026 at 11:50 PM Tushar Vyavahare
> > <tushar.vyavahare@intel.com> wrote:
> > >
> > > This series improves AF_XDP selftests by making timeout handling
> > > explicit and fixing sources of non-determinism in xsk timeout tests.
> > >
> > > Patch 1 introduces test_spec::poll_tmout and removes implicit
> > > dependence on RX UMEM setup state for timeout behavior.
> > >
> > > Patch 2 fixes thread harness sequencing by attaching XDP programs
> > > before worker startup, removing signal-based termination, and using
> > > barrier synchronization only for dual-thread runs.
> > >
> > > Patch 3 restores shared_umem after POLL_TXQ_FULL so test-local
> > > configuration does not leak into subsequent cases on shared-netdev
> > > runs.
> > >
> > > Together these changes make timeout handling easier to follow and
> > > improve selftest stability, especially on real NIC runs.
> >
> > net-next is closed, but in the meantime I'll review the series ASAP.
> >
> > BTW, another thing about selftests I had in my mind is that are you
> > planning to work on this [1]?
>
> This one is on me. I took your changes Jason and aligned ZC batching side
> to this behavior, followed by xskxceiver adjustment. I am planning to send
> this today EOD, however let's see how badly internal Sashiko will kick my
> ass.

Thanks. We'll see :)

Thanks,
Jason

>
> >
> > [1]: https://lore.kernel.org/all/20260520004244.55663-1-kerneljasonxing@gmail.com/
> >
> > Thanks,
> > Jason
> >
> > >
> > > Tushar Vyavahare (3):
> > >   selftests/xsk: make poll timeout mode explicit
> > >   selftests/xsk: fix timeout thread harness sequencing
> > >   selftests/xsk: restore shared_umem after POLL_TXQ_FULL
> > >
> > >  .../selftests/bpf/prog_tests/test_xsk.c       | 96 +++++++++++--------
> > >  .../selftests/bpf/prog_tests/test_xsk.h       |  2 +
> > >  2 files changed, 56 insertions(+), 42 deletions(-)
> > >
> > > --
> > > 2.43.0
> > >
> > >
> >

Re: [PATCH net-next 0/3] selftests/xsk: stabilize timeout test behavior

[Jakub Kicinski]

On Wed, 17 Jun 2026 11:43:14 +0200 Maciej Fijalkowski wrote:
> > On Tue, Jun 16, 2026 at 11:50 PM Tushar Vyavahare
> > <tushar.vyavahare@intel.com> wrote:  
> > >
> > > This series improves AF_XDP selftests by making timeout handling
> > > explicit and fixing sources of non-determinism in xsk timeout tests.
> > >
> > > Patch 1 introduces test_spec::poll_tmout and removes implicit
> > > dependence on RX UMEM setup state for timeout behavior.
> > >
> > > Patch 2 fixes thread harness sequencing by attaching XDP programs
> > > before worker startup, removing signal-based termination, and using
> > > barrier synchronization only for dual-thread runs.
> > >
> > > Patch 3 restores shared_umem after POLL_TXQ_FULL so test-local
> > > configuration does not leak into subsequent cases on shared-netdev
> > > runs.
> > >
> > > Together these changes make timeout handling easier to follow and
> > > improve selftest stability, especially on real NIC runs.  
> > 
> > net-next is closed, but in the meantime I'll review the series ASAP.
> > 
> > BTW, another thing about selftests I had in my mind is that are you
> > planning to work on this [1]?  
> 
> This one is on me. I took your changes Jason and aligned ZC batching side
> to this behavior, followed by xskxceiver adjustment. I am planning to send
> this today EOD, however let's see how badly internal Sashiko will kick my
> ass.

Hi Maciej, do you want these applied? If they help make the tests less
flaky I think that it's fine to take them during the merge window.

Re: [PATCH net-next 0/3] selftests/xsk: stabilize timeout test behavior

[Jason Xing]

On Tue, Jun 23, 2026 at 7:07 AM Jakub Kicinski <kuba@kernel.org> wrote:
>
> On Wed, 17 Jun 2026 11:43:14 +0200 Maciej Fijalkowski wrote:
> > > On Tue, Jun 16, 2026 at 11:50 PM Tushar Vyavahare
> > > <tushar.vyavahare@intel.com> wrote:
> > > >
> > > > This series improves AF_XDP selftests by making timeout handling
> > > > explicit and fixing sources of non-determinism in xsk timeout tests.
> > > >
> > > > Patch 1 introduces test_spec::poll_tmout and removes implicit
> > > > dependence on RX UMEM setup state for timeout behavior.
> > > >
> > > > Patch 2 fixes thread harness sequencing by attaching XDP programs
> > > > before worker startup, removing signal-based termination, and using
> > > > barrier synchronization only for dual-thread runs.
> > > >
> > > > Patch 3 restores shared_umem after POLL_TXQ_FULL so test-local
> > > > configuration does not leak into subsequent cases on shared-netdev
> > > > runs.
> > > >
> > > > Together these changes make timeout handling easier to follow and
> > > > improve selftest stability, especially on real NIC runs.
> > >
> > > net-next is closed, but in the meantime I'll review the series ASAP.
> > >
> > > BTW, another thing about selftests I had in my mind is that are you
> > > planning to work on this [1]?
> >
> > This one is on me. I took your changes Jason and aligned ZC batching side
> > to this behavior, followed by xskxceiver adjustment. I am planning to send
> > this today EOD, however let's see how badly internal Sashiko will kick my
> > ass.
>
> Hi Maciej, do you want these applied? If they help make the tests less
> flaky I think that it's fine to take them during the merge window.

I'm not Maciej, and my take on this overall looks fine to me. Sorry
for the delay (I've been too busy to review patches during this
period).

Reviewed-by: Jason Xing <kerneljasonxing@gmail.com>

Thanks,
Jason

Re: [PATCH net-next 0/3] selftests/xsk: stabilize timeout test behavior

[Maciej Fijalkowski]

On Mon, Jun 22, 2026 at 04:07:06PM -0700, Jakub Kicinski wrote:
> On Wed, 17 Jun 2026 11:43:14 +0200 Maciej Fijalkowski wrote:
> > > On Tue, Jun 16, 2026 at 11:50 PM Tushar Vyavahare
> > > <tushar.vyavahare@intel.com> wrote:  
> > > >
> > > > This series improves AF_XDP selftests by making timeout handling
> > > > explicit and fixing sources of non-determinism in xsk timeout tests.
> > > >
> > > > Patch 1 introduces test_spec::poll_tmout and removes implicit
> > > > dependence on RX UMEM setup state for timeout behavior.
> > > >
> > > > Patch 2 fixes thread harness sequencing by attaching XDP programs
> > > > before worker startup, removing signal-based termination, and using
> > > > barrier synchronization only for dual-thread runs.
> > > >
> > > > Patch 3 restores shared_umem after POLL_TXQ_FULL so test-local
> > > > configuration does not leak into subsequent cases on shared-netdev
> > > > runs.
> > > >
> > > > Together these changes make timeout handling easier to follow and
> > > > improve selftest stability, especially on real NIC runs.  
> > > 
> > > net-next is closed, but in the meantime I'll review the series ASAP.
> > > 
> > > BTW, another thing about selftests I had in my mind is that are you
> > > planning to work on this [1]?  
> > 
> > This one is on me. I took your changes Jason and aligned ZC batching side
> > to this behavior, followed by xskxceiver adjustment. I am planning to send
> > this today EOD, however let's see how badly internal Sashiko will kick my
> > ass.
> 
> Hi Maciej, do you want these applied? If they help make the tests less
> flaky I think that it's fine to take them during the merge window.

Hi Jakub,

last refactor from Tushar broke BIDIRECTIONAL test case when HW is test
target, but not on veth, so let me test these changes locally and then get
back to you.

BPF CI runs xskxceiver on veth so this has not been caught. Seems my/our
focus should be to enable xskxceiver HW tests on any kind of
environment/infrastructure.

Gonna get back to you by the EOD.
Maciej