From: Jakub Kicinski <kuba@kernel.org>
To: Mohsin Bashir <mohsin.bashr@gmail.com>
Cc: netdev@vger.kernel.org, andrew+netdev@lunn.ch,
davem@davemloft.net, edumazet@google.com, pabeni@redhat.com,
shuah@kernel.org, linux-kselftest@vger.kernel.org
Subject: Re: [PATCH net-next V2] selftests: drv-net: Test queue stall upon reconfig
Date: Sat, 13 Jun 2026 09:20:02 -0700 [thread overview]
Message-ID: <20260613092002.72fc7906@kernel.org> (raw)
In-Reply-To: <20260613014855.1717712-1-mohsin.bashr@gmail.com>
On Fri, 12 Jun 2026 18:48:54 -0700 Mohsin Bashir wrote:
> From: Mohsin Bashir <hmohsin@meta.com>
>
> Add a reconfig_tx_stall test that detects the possibility of a TX stall
> after ring reconfiguration. The key observation is that drivers using
> netif_tx_start_all_queues() are prone to experiencing a stall when
> reconfiguration completes compared to drivers using
> netif_tx_wake_all_queues(). start_all_queues only clears DRV_XOFF, while
> wake_all_queues also calls __netif_schedule() to kick the qdisc. Without
> the kick, qdisc backlog present at reconfig time can stay stuck until a
> new trigger is issued.
>
> The test caps the TX ring at 64 entries so it fills quickly, then
> installs FQ on a target TX queue and sends UDP packets with SO_TXTIME
> scheduled in the future. With napi_defer_hard_irqs slowing completions,
> the small ring can fill when FQ releases the burst, leaving requeued
> qdisc backlog with no FQ timer to rescue it. A subsequent ring reconfig
> must wake the queues to drain the backlog. Simply starting the queues can
> leave it stuck.
>
> On host with problematic driver:
> Sent 128 SO_TXTIME packets (+100ms)
> Sent 128 SO_TXTIME packets (+200ms)
> Backlog before reconfig: 52632 bytes
> Check| At /root/ksft-net-drv/./drivers/net/ring_reconfig.py, ...
> Check| ksft_eq(0, backlog,
> Check failed 0 != 52632 qdisc backlog stuck on queue 1 after ring reconfig
> not ok 3 ring_reconfig.reconfig_tx_stall
>
> On host with fixed driver:
> Sent 128 SO_TXTIME packets (+100ms)
> Sent 128 SO_TXTIME packets (+200ms)
> Backlog before reconfig: 76024 bytes
> ok 3 ring_reconfig.reconfig_tx_stall
>
> Signed-off-by: Mohsin Bashir <hmohsin@meta.com>
> Signed-off-by: Jakub Kicinski <kuba@kernel.org>
pylint is not on board:
+tools/testing/selftests/drivers/net/ring_reconfig.py:169:37: W1514: Using open without explicitly specifying an encoding (unspecified-encoding)
+tools/testing/selftests/drivers/net/ring_reconfig.py:162:17: W0613: Unused argument 'cfg' (unused-argument)
+tools/testing/selftests/drivers/net/ring_reconfig.py:253:0: C0116: Missing function or method docstring (missing-function-docstring)
> diff --git a/tools/testing/selftests/drivers/net/config b/tools/testing/selftests/drivers/net/config
> index 617de8aaf551..1ef07fae74c1 100644
> --- a/tools/testing/selftests/drivers/net/config
> +++ b/tools/testing/selftests/drivers/net/config
> @@ -4,6 +4,10 @@ CONFIG_DEBUG_INFO_BTF_MODULES=n
> CONFIG_INET_PSP=y
> CONFIG_IPV6=y
> CONFIG_MACSEC=m
> +CONFIG_NET_ACT_SKBEDIT=m
> +CONFIG_NET_CLS_ACT=y
> +CONFIG_NET_CLS_FLOWER=m
> +CONFIG_NET_CLS_MATCHALL=m
> CONFIG_NETCONSOLE=m
> CONFIG_NETCONSOLE_DYNAMIC=y
> CONFIG_NETCONSOLE_EXTENDED_LOG=y
> diff --git a/tools/testing/selftests/drivers/net/ring_reconfig.py b/tools/testing/selftests/drivers/net/ring_reconfig.py
> index f9530a8b0856..11491a0b7013 100755
> --- a/tools/testing/selftests/drivers/net/ring_reconfig.py
> +++ b/tools/testing/selftests/drivers/net/ring_reconfig.py
> @@ -5,10 +5,18 @@
> Test channel and ring size configuration via ethtool (-L / -G).
> """
>
> +import socket
> +import struct
> +import time
> +
> from lib.py import ksft_run, ksft_exit, ksft_pr
> from lib.py import ksft_eq
> +from lib.py import KsftSkipEx
> from lib.py import NetDrvEpEnv, EthtoolFamily, GenerateTraffic
> -from lib.py import defer, NlError
> +from lib.py import cmd, defer, rand_port, tc, NlError
> +
> +# Added in Python 3.13; fallback to 61 for x86/ARM/MIPS
> +SO_TXTIME = getattr(socket, "SO_TXTIME", 61)
>
>
> def channels(cfg) -> None:
> @@ -151,6 +159,169 @@ def ringparam(cfg) -> None:
> GenerateTraffic(cfg).wait_pkts_and_stop(10000)
>
>
> +def _write_sysfs(cfg, path, val):
> + with open(path, "r", encoding="utf-8") as fp:
> + orig_val = fp.read().strip()
> + if str(val) == orig_val:
> + return
> + with open(path, "w", encoding="utf-8") as fp:
> + fp.write(str(val))
> + defer(lambda p=path, v=orig_val: open(p, "w").write(v))
> +
> +
> +def _get_mq_handle(cfg):
> + qdiscs = tc(f"qdisc show dev {cfg.ifname}", json=True)
> + for q in qdiscs:
> + if q.get("kind") == "mq":
> + return q["handle"]
> + raise KsftSkipEx(f"no mq qdisc found on {cfg.ifname}")
> +
> +
> +def _get_qdisc_backlog(cfg, queue, mq_handle):
> + qdiscs = tc(f"-s qdisc show dev {cfg.ifname}", json=True)
> + target_parent = f"{mq_handle}{queue + 1:x}"
> + for q in qdiscs:
> + if q.get("parent", "") == target_parent:
> + return q.get("backlog")
> + return None
> +
> +
> +def _setup_fq_qdisc(cfg, mq_handle, port, target_queue, other_queue):
> + mq_child_parent = f"{mq_handle}{target_queue + 1:x}"
> +
> + # Save the original child qdisc to restore after test
> + qdiscs = tc(f"qdisc show dev {cfg.ifname}", json=True)
> + default_qdisc = cmd("sysctl -n net.core.default_qdisc").stdout.strip()
> + orig_kind = default_qdisc
> + for q in qdiscs:
> + if q.get("parent", "") == mq_child_parent:
> + orig_kind = q.get("kind", default_qdisc)
> + break
> + try:
> + tc(f"qdisc replace dev {cfg.ifname} parent {mq_child_parent} fq")
> + except Exception as exc:
> + raise KsftSkipEx("fq not available (CONFIG_NET_SCH_FQ)") from exc
> + defer(tc,
> + f"qdisc replace dev {cfg.ifname} parent {mq_child_parent} {orig_kind}")
> +
> + qdisc_j = tc(f"qdisc show dev {cfg.ifname}", json=True)
> + has_clsact = any(q['kind'] == 'clsact' for q in qdisc_j)
> + if not has_clsact:
> + tc(f"qdisc add dev {cfg.ifname} clsact")
> + defer(tc, f"qdisc del dev {cfg.ifname} clsact")
> +
> + proto = "ipv6" if int(cfg.addr_ipver) == 6 else "ip"
> + try:
> + tc(f"filter add dev {cfg.ifname} egress protocol {proto} "
> + f"pref 1 flower ip_proto udp dst_port {port} "
> + f"action skbedit queue_mapping {target_queue}")
> + except Exception as exc:
> + raise KsftSkipEx("tc flower/act_skbedit not available") from exc
> + defer(tc, f"filter del dev {cfg.ifname} egress pref 1")
> +
> + tc(f"filter add dev {cfg.ifname} egress pref 100 "
> + f"matchall action skbedit queue_mapping {other_queue}")
> + defer(tc, f"filter del dev {cfg.ifname} egress pref 100")
> +
> +
> +def _create_sotxtime_socket(cfg):
> + sock = socket.socket(socket.AF_INET6 if cfg.addr_ipver == "6"
> + else socket.AF_INET, socket.SOCK_DGRAM)
> + try:
> + sock.setsockopt(socket.SOL_SOCKET, SO_TXTIME, struct.pack("Ii", 1, 0))
> + except OSError as exc:
> + sock.close()
> + raise KsftSkipEx("SO_TXTIME not supported") from exc
> + sock.setsockopt(socket.SOL_SOCKET, socket.SO_BINDTODEVICE,
> + cfg.ifname.encode())
> + return sock
> +
> +
> +def _send_sotxtime_burst(sock, addr, port, count, delay_ns, ipver):
> + payload = b'\x00' * 1400
> + txtime_ns = time.clock_gettime_ns(time.CLOCK_MONOTONIC) + delay_ns
> +
> + ancdata = [(socket.SOL_SOCKET, SO_TXTIME, struct.pack("Q", txtime_ns))]
> + if int(ipver) == 6:
> + dest = (addr, port, 0, 0)
> + else:
> + dest = (addr, port)
> + for _ in range(count):
> + sock.sendmsg([payload], ancdata, 0, dest)
> +
> +
> +def reconfig_tx_stall(cfg) -> None:
> + target_queue = 1
> + other_queue = 0
> +
> + ehdr = {'header': {'dev-index': cfg.ifindex}}
> + chans = cfg.eth.channels_get(ehdr)
> +
> + if 'combined-max' not in chans:
> + raise KsftSkipEx("device does not support combined channels")
> + if chans['combined-count'] < 2:
> + raise KsftSkipEx("need at least 2 combined channels")
> +
> + rings = cfg.eth.rings_get(ehdr)
> + if 'rx' not in rings or 'tx' not in rings:
> + raise KsftSkipEx("device does not expose rx/tx ring params")
> + tx_cur = rings['tx']
> + if tx_cur <= 64:
> + raise KsftSkipEx("tx ring size already at minimum")
> + defer(cfg.eth.rings_set, ehdr | {'tx': tx_cur})
> +
> + tx_min = 64
> + cfg.eth.rings_set(ehdr | {'tx': tx_min})
> +
> + # Slow completions so the ring stays full after FQ releases packets
> + napi_defer = f"/sys/class/net/{cfg.ifname}/napi_defer_hard_irqs"
> + gro_timeout = f"/sys/class/net/{cfg.ifname}/gro_flush_timeout"
> + _write_sysfs(cfg, napi_defer, 100)
> + _write_sysfs(cfg, gro_timeout, 1000000000)
> +
> + mq_handle = _get_mq_handle(cfg)
> + port = rand_port()
> + _setup_fq_qdisc(cfg, mq_handle, port, target_queue, other_queue)
> +
> + sock = _create_sotxtime_socket(cfg)
> + defer(sock.close)
> +
> + pkt_count = tx_min * 2
> +
> + for delay_ms in [100, 200, 500]:
> + delay_ns = delay_ms * 1_000_000
> + _send_sotxtime_burst(sock, cfg.remote_addr, port, pkt_count,
> + delay_ns, cfg.addr_ipver)
> + ksft_pr(f"Sent {pkt_count} SO_TXTIME packets (+{delay_ms}ms)")
> + time.sleep(delay_ms / 1000 + 0.3)
> +
> + backlog = _get_qdisc_backlog(cfg, target_queue, mq_handle)
> + if backlog:
> + break
> + else:
> + raise KsftSkipEx("failed to build qdisc backlog")
> +
> + ksft_pr(f"Backlog before reconfig: {backlog} bytes")
> +
> + # Trigger ring reconfig — driver should call wake, not just start
> + cfg.eth.rings_set(ehdr | {'tx': tx_cur})
> +
> + # Let completions proceed normally
> + _write_sysfs(cfg, napi_defer, 0)
> + _write_sysfs(cfg, gro_timeout, 0)
> +
> + # Poll for backlog to drain
> + for _ in range(100):
> + backlog = _get_qdisc_backlog(cfg, target_queue, mq_handle)
> + if not backlog:
> + break
> + time.sleep(0.1)
> +
> + ksft_eq(0, backlog,
> + comment=f"qdisc backlog stuck on queue {target_queue} "
> + f"after ring reconfig")
> +
> +
> def main() -> None:
> """ Ksft boiler plate main """
>
> @@ -158,7 +329,8 @@ def main() -> None:
the NetDrvEpEnv() setup needs to ask for 2+ queues, otherwise this
fails in netdevsim mode:
# ok 3 ring_reconfig.reconfig_tx_stall # SKIP need at least 2 combined channels
> cfg.eth = EthtoolFamily()
>
> ksft_run([channels,
> - ringparam],
> + ringparam,
> + reconfig_tx_stall],
> args=(cfg, ))
> ksft_exit()
>
prev parent reply other threads:[~2026-06-13 16:20 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-06-13 1:48 [PATCH net-next V2] selftests: drv-net: Test queue stall upon reconfig Mohsin Bashir
2026-06-13 16:20 ` Jakub Kicinski [this message]
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=20260613092002.72fc7906@kernel.org \
--to=kuba@kernel.org \
--cc=andrew+netdev@lunn.ch \
--cc=davem@davemloft.net \
--cc=edumazet@google.com \
--cc=linux-kselftest@vger.kernel.org \
--cc=mohsin.bashr@gmail.com \
--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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox