From 0163250039c3a861f922a293f6108880b2d4516f Mon Sep 17 00:00:00 2001
From: Gal Pressman <gal@nvidia.com>
Date: Mon, 24 Feb 2025 19:44:15 +0200
Subject: [PATCH 01/16] selftests: drv-net: Make rand_port() get a port more
 reliably

Instead of guessing a port and checking whether it's available, get an
available port from the OS.

Reviewed-by: Nimrod Oren <noren@nvidia.com>
Signed-off-by: Gal Pressman <gal@nvidia.com>
Link: https://patch.msgid.link/20250224174416.499070-4-gal@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 tools/testing/selftests/net/lib/py/utils.py | 15 ++++-----------
 1 file changed, 4 insertions(+), 11 deletions(-)

diff --git a/tools/testing/selftests/net/lib/py/utils.py b/tools/testing/selftests/net/lib/py/utils.py
index d879700ef2b9..a6af97c7e283 100644
--- a/tools/testing/selftests/net/lib/py/utils.py
+++ b/tools/testing/selftests/net/lib/py/utils.py
@@ -187,18 +187,11 @@ def ethtool(args, json=None, ns=None, host=None):
 
 def rand_port():
     """
-    Get a random unprivileged port, try to make sure it's not already used.
+    Get a random unprivileged port.
     """
-    for _ in range(1000):
-        port = random.randint(10000, 65535)
-        try:
-            with socket.socket(socket.AF_INET6, socket.SOCK_STREAM) as s:
-                s.bind(("", port))
-            return port
-        except OSError as e:
-            if e.errno != errno.EADDRINUSE:
-                raise
-    raise Exception("Can't find any free unprivileged port")
+    with socket.socket(socket.AF_INET6, socket.SOCK_STREAM) as s:
+        s.bind(("", 0))
+        return s.getsockname()[1]
 
 
 def wait_port_listen(port, proto="tcp", ns=None, host=None, sleep=0.005, deadline=5):
-- 
2.51.0


From da87cabaf87702d43a016d255f11be5379892b6a Mon Sep 17 00:00:00 2001
From: Gal Pressman <gal@nvidia.com>
Date: Mon, 24 Feb 2025 19:44:16 +0200
Subject: [PATCH 02/16] selftests: drv-net-hw: Add a test for symmetric RSS
 hash

Add a selftest that verifies symmetric RSS hash is working as intended.
The test runs iterations of traffic, swapping the src/dst UDP ports, and
verifies that the same RX queue is receiving the traffic in both cases.

Reviewed-by: Nimrod Oren <noren@nvidia.com>
Signed-off-by: Gal Pressman <gal@nvidia.com>
Link: https://patch.msgid.link/20250224174416.499070-5-gal@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 .../testing/selftests/drivers/net/hw/Makefile |  1 +
 .../drivers/net/hw/rss_input_xfrm.py          | 87 +++++++++++++++++++
 tools/testing/selftests/net/lib/py/utils.py   |  4 +-
 3 files changed, 90 insertions(+), 2 deletions(-)
 create mode 100755 tools/testing/selftests/drivers/net/hw/rss_input_xfrm.py

diff --git a/tools/testing/selftests/drivers/net/hw/Makefile b/tools/testing/selftests/drivers/net/hw/Makefile
index ae783e18be83..b2be14d0cc56 100644
--- a/tools/testing/selftests/drivers/net/hw/Makefile
+++ b/tools/testing/selftests/drivers/net/hw/Makefile
@@ -15,6 +15,7 @@ TEST_PROGS = \
 	nic_performance.py \
 	pp_alloc_fail.py \
 	rss_ctx.py \
+	rss_input_xfrm.py \
 	tso.py \
 	#
 
diff --git a/tools/testing/selftests/drivers/net/hw/rss_input_xfrm.py b/tools/testing/selftests/drivers/net/hw/rss_input_xfrm.py
new file mode 100755
index 000000000000..53bb08cc29ec
--- /dev/null
+++ b/tools/testing/selftests/drivers/net/hw/rss_input_xfrm.py
@@ -0,0 +1,87 @@
+#!/usr/bin/env python3
+# SPDX-License-Identifier: GPL-2.0
+
+import multiprocessing
+import socket
+from lib.py import ksft_run, ksft_exit, ksft_eq, ksft_ge, cmd, fd_read_timeout
+from lib.py import NetDrvEpEnv
+from lib.py import EthtoolFamily, NetdevFamily
+from lib.py import KsftSkipEx, KsftFailEx
+from lib.py import rand_port
+
+
+def traffic(cfg, local_port, remote_port, ipver):
+    af_inet = socket.AF_INET if ipver == "4" else socket.AF_INET6
+    sock = socket.socket(af_inet, socket.SOCK_DGRAM)
+    sock.bind(("", local_port))
+    sock.connect((cfg.remote_addr_v[ipver], remote_port))
+    tgt = f"{ipver}:[{cfg.addr_v[ipver]}]:{local_port},sourceport={remote_port}"
+    cmd("echo a | socat - UDP" + tgt, host=cfg.remote)
+    fd_read_timeout(sock.fileno(), 5)
+    return sock.getsockopt(socket.SOL_SOCKET, socket.SO_INCOMING_CPU)
+
+
+def test_rss_input_xfrm(cfg, ipver):
+    """
+    Test symmetric input_xfrm.
+    If symmetric RSS hash is configured, send traffic twice, swapping the
+    src/dst UDP ports, and verify that the same queue is receiving the traffic
+    in both cases (IPs are constant).
+    """
+
+    if multiprocessing.cpu_count() < 2:
+        raise KsftSkipEx("Need at least two CPUs to test symmetric RSS hash")
+
+    input_xfrm = cfg.ethnl.rss_get(
+        {'header': {'dev-name': cfg.ifname}}).get('input_xfrm')
+
+    # Check for symmetric xor/or-xor
+    if not input_xfrm or (input_xfrm != 1 and input_xfrm != 2):
+        raise KsftSkipEx("Symmetric RSS hash not requested")
+
+    cpus = set()
+    successful = 0
+    for _ in range(100):
+        try:
+            port1 = rand_port(socket.SOCK_DGRAM)
+            port2 = rand_port(socket.SOCK_DGRAM)
+            cpu1 = traffic(cfg, port1, port2, ipver)
+            cpu2 = traffic(cfg, port2, port1, ipver)
+            cpus.update([cpu1, cpu2])
+            ksft_eq(
+                cpu1, cpu2, comment=f"Received traffic on different cpus with ports ({port1 = }, {port2 = }) while symmetric hash is configured")
+
+            successful += 1
+            if successful == 10:
+                break
+        except:
+            continue
+    else:
+        raise KsftFailEx("Failed to run traffic")
+
+    ksft_ge(len(cpus), 2,
+            comment=f"Received traffic on less than two cpus {cpus = }")
+
+
+def test_rss_input_xfrm_ipv4(cfg):
+    cfg.require_ipver("4")
+    test_rss_input_xfrm(cfg, "4")
+
+
+def test_rss_input_xfrm_ipv6(cfg):
+    cfg.require_ipver("6")
+    test_rss_input_xfrm(cfg, "6")
+
+
+def main() -> None:
+    with NetDrvEpEnv(__file__, nsim_test=False) as cfg:
+        cfg.ethnl = EthtoolFamily()
+        cfg.netdevnl = NetdevFamily()
+
+        ksft_run([test_rss_input_xfrm_ipv4, test_rss_input_xfrm_ipv6],
+                 args=(cfg, ))
+    ksft_exit()
+
+
+if __name__ == "__main__":
+    main()
diff --git a/tools/testing/selftests/net/lib/py/utils.py b/tools/testing/selftests/net/lib/py/utils.py
index a6af97c7e283..34470d65d871 100644
--- a/tools/testing/selftests/net/lib/py/utils.py
+++ b/tools/testing/selftests/net/lib/py/utils.py
@@ -185,11 +185,11 @@ def ethtool(args, json=None, ns=None, host=None):
     return tool('ethtool', args, json=json, ns=ns, host=host)
 
 
-def rand_port():
+def rand_port(type=socket.SOCK_STREAM):
     """
     Get a random unprivileged port.
     """
-    with socket.socket(socket.AF_INET6, socket.SOCK_STREAM) as s:
+    with socket.socket(socket.AF_INET6, type) as s:
         s.bind(("", 0))
         return s.getsockname()[1]
 
-- 
2.51.0


From 8fa19c2c69fbf1f615e84d24d75dcf001ea91ccb Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Mon, 24 Feb 2025 15:05:06 +0100
Subject: [PATCH 03/16] net: wangxun: fix LIBWX dependencies

Selecting LIBWX requires that its dependencies are met first:

WARNING: unmet direct dependencies detected for LIBWX
  Depends on [m]: NETDEVICES [=y] && ETHERNET [=y] && NET_VENDOR_WANGXUN [=y] && PTP_1588_CLOCK_OPTIONAL [=m]
  Selected by [y]:
  - TXGBE [=y] && NETDEVICES [=y] && ETHERNET [=y] && NET_VENDOR_WANGXUN [=y] && PCI [=y] && COMMON_CLK [=y] && I2C_DESIGNWARE_PLATFORM [=y]
ld.lld-21: error: undefined symbol: ptp_schedule_worker
>>> referenced by wx_ptp.c:747 (/home/arnd/arm-soc/drivers/net/ethernet/wangxun/libwx/wx_ptp.c:747)
>>>               drivers/net/ethernet/wangxun/libwx/wx_ptp.o:(wx_ptp_reset) in archive vmlinux.a

Add the smae dependency on PTP_1588_CLOCK_OPTIONAL to the two driver
using this library module.

Fixes: 06e75161b9d4 ("net: wangxun: Add support for PTP clock")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Reviewed-by: Vadim Fedorenko <vadim.fedorenko@linux.dev>
Link: https://patch.msgid.link/20250224140516.1168214-1-arnd@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/wangxun/Kconfig | 2 ++
 1 file changed, 2 insertions(+)

diff --git a/drivers/net/ethernet/wangxun/Kconfig b/drivers/net/ethernet/wangxun/Kconfig
index 6b60173fe1f5..47e3e8434b9e 100644
--- a/drivers/net/ethernet/wangxun/Kconfig
+++ b/drivers/net/ethernet/wangxun/Kconfig
@@ -26,6 +26,7 @@ config LIBWX
 config NGBE
 	tristate "Wangxun(R) GbE PCI Express adapters support"
 	depends on PCI
+	depends on PTP_1588_CLOCK_OPTIONAL
 	select LIBWX
 	select PHYLINK
 	help
@@ -43,6 +44,7 @@ config TXGBE
 	depends on PCI
 	depends on COMMON_CLK
 	depends on I2C_DESIGNWARE_PLATFORM
+	depends on PTP_1588_CLOCK_OPTIONAL
 	select MARVELL_10G_PHY
 	select REGMAP
 	select PHYLINK
-- 
2.51.0


From 91c8d8e4b7a38dc099b26e14b22f814ca4e75089 Mon Sep 17 00:00:00 2001
From: John Daley <johndale@cisco.com>
Date: Mon, 24 Feb 2025 15:43:50 -0800
Subject: [PATCH 04/16] enic: add dependency on Page Pool

Driver was not configured to select page_pool, causing a compile error
if page pool module was not already selected.

Reported-by: kernel test robot <lkp@intel.com>
Closes: https://lore.kernel.org/oe-kbuild-all/202502211253.3XRosM9I-lkp@intel.com/
Fixes: d24cb52b2d8a ("enic: Use the Page Pool API for RX")
Reviewed-by: Nelson Escobar <neescoba@cisco.com>
Signed-off-by: John Daley <johndale@cisco.com>
Reviewed-by: Joe Damato <jdamato@fastly.com>
Link: https://patch.msgid.link/20250224234350.23157-2-johndale@cisco.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/cisco/enic/Kconfig | 1 +
 1 file changed, 1 insertion(+)

diff --git a/drivers/net/ethernet/cisco/enic/Kconfig b/drivers/net/ethernet/cisco/enic/Kconfig
index ad80c0fa96a6..96709875fe4f 100644
--- a/drivers/net/ethernet/cisco/enic/Kconfig
+++ b/drivers/net/ethernet/cisco/enic/Kconfig
@@ -6,5 +6,6 @@
 config ENIC
 	tristate "Cisco VIC Ethernet NIC Support"
 	depends on PCI
+	select PAGE_POOL
 	help
 	  This enables the support for the Cisco VIC Ethernet card.
-- 
2.51.0


From 3ba075278c11cdb19e2dbb80362042f1b0c08f74 Mon Sep 17 00:00:00 2001
From: Eric Dumazet <edumazet@google.com>
Date: Tue, 25 Feb 2025 17:10:48 +0000
Subject: [PATCH 05/16] tcp: be less liberal in TSEcr received while in
 SYN_RECV state

Yong-Hao Zou mentioned that linux was not strict as other OS in 3WHS,
for flows using TCP TS option (RFC 7323)

As hinted by an old comment in tcp_check_req(),
we can check the TSEcr value in the incoming packet corresponds
to one of the SYNACK TSval values we have sent.

In this patch, I record the oldest and most recent values
that SYNACK packets have used.

Send a challenge ACK if we receive a TSEcr outside
of this range, and increase a new SNMP counter.

nstat -az | grep TSEcrRejected
TcpExtTSEcrRejected            0                  0.0

Due to TCP fastopen implementation, do not apply yet these checks
for fastopen flows.

v2: No longer use req->num_timeout, but treq->snt_tsval_first
    to detect when first SYNACK is prepared. This means
    we make sure to not send an initial zero TSval.
    Make sure MPTCP and TCP selftests are passing.
    Change MIB name to TcpExtTSEcrRejected

v1: https://lore.kernel.org/netdev/CADVnQykD8i4ArpSZaPKaoNxLJ2if2ts9m4As+=Jvdkrgx1qMHw@mail.gmail.com/T/

Reported-by: Yong-Hao Zou <yonghaoz1994@gmail.com>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Matthieu Baerts (NGI0) <matttbe@kernel.org>
Reviewed-by: Neal Cardwell <ncardwell@google.com>
Reviewed-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Link: https://patch.msgid.link/20250225171048.3105061-1-edumazet@google.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 .../networking/net_cachelines/snmp.rst        |  1 +
 include/linux/tcp.h                           |  2 ++
 include/uapi/linux/snmp.h                     |  1 +
 net/ipv4/proc.c                               |  1 +
 net/ipv4/syncookies.c                         |  1 +
 net/ipv4/tcp_input.c                          |  1 +
 net/ipv4/tcp_minisocks.c                      | 26 +++++++++++--------
 net/ipv4/tcp_output.c                         |  6 +++++
 8 files changed, 28 insertions(+), 11 deletions(-)

diff --git a/Documentation/networking/net_cachelines/snmp.rst b/Documentation/networking/net_cachelines/snmp.rst
index 90ca2d92547d..bc96efc92cf5 100644
--- a/Documentation/networking/net_cachelines/snmp.rst
+++ b/Documentation/networking/net_cachelines/snmp.rst
@@ -36,6 +36,7 @@ unsigned_long  LINUX_MIB_TIMEWAITRECYCLED
 unsigned_long  LINUX_MIB_TIMEWAITKILLED
 unsigned_long  LINUX_MIB_PAWSACTIVEREJECTED
 unsigned_long  LINUX_MIB_PAWSESTABREJECTED
+unsigned_long  LINUX_MIB_TSECR_REJECTED
 unsigned_long  LINUX_MIB_DELAYEDACKLOST
 unsigned_long  LINUX_MIB_LISTENOVERFLOWS
 unsigned_long  LINUX_MIB_LISTENDROPS
diff --git a/include/linux/tcp.h b/include/linux/tcp.h
index f88daaa76d83..159b2c59eb62 100644
--- a/include/linux/tcp.h
+++ b/include/linux/tcp.h
@@ -160,6 +160,8 @@ struct tcp_request_sock {
 	u32				rcv_isn;
 	u32				snt_isn;
 	u32				ts_off;
+	u32				snt_tsval_first;
+	u32				snt_tsval_last;
 	u32				last_oow_ack_time; /* last SYNACK */
 	u32				rcv_nxt; /* the ack # by SYNACK. For
 						  * FastOpen it's the seq#
diff --git a/include/uapi/linux/snmp.h b/include/uapi/linux/snmp.h
index 848c7784e684..eb9fb776fdc3 100644
--- a/include/uapi/linux/snmp.h
+++ b/include/uapi/linux/snmp.h
@@ -186,6 +186,7 @@ enum
 	LINUX_MIB_TIMEWAITKILLED,		/* TimeWaitKilled */
 	LINUX_MIB_PAWSACTIVEREJECTED,		/* PAWSActiveRejected */
 	LINUX_MIB_PAWSESTABREJECTED,		/* PAWSEstabRejected */
+	LINUX_MIB_TSECRREJECTED,		/* TSEcrRejected */
 	LINUX_MIB_PAWS_OLD_ACK,			/* PAWSOldAck */
 	LINUX_MIB_DELAYEDACKS,			/* DelayedACKs */
 	LINUX_MIB_DELAYEDACKLOCKED,		/* DelayedACKLocked */
diff --git a/net/ipv4/proc.c b/net/ipv4/proc.c
index affd21a0f572..10cbeb76c274 100644
--- a/net/ipv4/proc.c
+++ b/net/ipv4/proc.c
@@ -189,6 +189,7 @@ static const struct snmp_mib snmp4_net_list[] = {
 	SNMP_MIB_ITEM("TWKilled", LINUX_MIB_TIMEWAITKILLED),
 	SNMP_MIB_ITEM("PAWSActive", LINUX_MIB_PAWSACTIVEREJECTED),
 	SNMP_MIB_ITEM("PAWSEstab", LINUX_MIB_PAWSESTABREJECTED),
+	SNMP_MIB_ITEM("TSEcrRejected", LINUX_MIB_TSECRREJECTED),
 	SNMP_MIB_ITEM("PAWSOldAck", LINUX_MIB_PAWS_OLD_ACK),
 	SNMP_MIB_ITEM("DelayedACKs", LINUX_MIB_DELAYEDACKS),
 	SNMP_MIB_ITEM("DelayedACKLocked", LINUX_MIB_DELAYEDACKLOCKED),
diff --git a/net/ipv4/syncookies.c b/net/ipv4/syncookies.c
index 26816b876dd8..5459a78b9809 100644
--- a/net/ipv4/syncookies.c
+++ b/net/ipv4/syncookies.c
@@ -279,6 +279,7 @@ static int cookie_tcp_reqsk_init(struct sock *sk, struct sk_buff *skb,
 		ireq->smc_ok = 0;
 
 	treq->snt_synack = 0;
+	treq->snt_tsval_first = 0;
 	treq->tfo_listener = false;
 	treq->txhash = net_tx_rndhash();
 	treq->rcv_isn = ntohl(th->seq) - 1;
diff --git a/net/ipv4/tcp_input.c b/net/ipv4/tcp_input.c
index 217a8747a79b..d22ad553b45b 100644
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -7081,6 +7081,7 @@ static void tcp_openreq_init(struct request_sock *req,
 	tcp_rsk(req)->rcv_isn = TCP_SKB_CB(skb)->seq;
 	tcp_rsk(req)->rcv_nxt = TCP_SKB_CB(skb)->seq + 1;
 	tcp_rsk(req)->snt_synack = 0;
+	tcp_rsk(req)->snt_tsval_first = 0;
 	tcp_rsk(req)->last_oow_ack_time = 0;
 	req->mss = rx_opt->mss_clamp;
 	req->ts_recent = rx_opt->saw_tstamp ? rx_opt->rcv_tsval : 0;
diff --git a/net/ipv4/tcp_minisocks.c b/net/ipv4/tcp_minisocks.c
index 1eccc518b957..4f87406ddbcd 100644
--- a/net/ipv4/tcp_minisocks.c
+++ b/net/ipv4/tcp_minisocks.c
@@ -663,6 +663,7 @@ struct sock *tcp_check_req(struct sock *sk, struct sk_buff *skb,
 	struct sock *child;
 	const struct tcphdr *th = tcp_hdr(skb);
 	__be32 flg = tcp_flag_word(th) & (TCP_FLAG_RST|TCP_FLAG_SYN|TCP_FLAG_ACK);
+	bool tsecr_reject = false;
 	bool paws_reject = false;
 	bool own_req;
 
@@ -672,8 +673,13 @@ struct sock *tcp_check_req(struct sock *sk, struct sk_buff *skb,
 
 		if (tmp_opt.saw_tstamp) {
 			tmp_opt.ts_recent = READ_ONCE(req->ts_recent);
-			if (tmp_opt.rcv_tsecr)
+			if (tmp_opt.rcv_tsecr) {
+				if (inet_rsk(req)->tstamp_ok && !fastopen)
+					tsecr_reject = !between(tmp_opt.rcv_tsecr,
+							tcp_rsk(req)->snt_tsval_first,
+							READ_ONCE(tcp_rsk(req)->snt_tsval_last));
 				tmp_opt.rcv_tsecr -= tcp_rsk(req)->ts_off;
+			}
 			/* We do not store true stamp, but it is not required,
 			 * it can be estimated (approximately)
 			 * from another data.
@@ -788,18 +794,14 @@ struct sock *tcp_check_req(struct sock *sk, struct sk_buff *skb,
 	     tcp_rsk(req)->snt_isn + 1))
 		return sk;
 
-	/* Also, it would be not so bad idea to check rcv_tsecr, which
-	 * is essentially ACK extension and too early or too late values
-	 * should cause reset in unsynchronized states.
-	 */
-
 	/* RFC793: "first check sequence number". */
 
-	if (paws_reject || !tcp_in_window(TCP_SKB_CB(skb)->seq,
-					  TCP_SKB_CB(skb)->end_seq,
-					  tcp_rsk(req)->rcv_nxt,
-					  tcp_rsk(req)->rcv_nxt +
-					  tcp_synack_window(req))) {
+	if (paws_reject || tsecr_reject ||
+	    !tcp_in_window(TCP_SKB_CB(skb)->seq,
+			   TCP_SKB_CB(skb)->end_seq,
+			   tcp_rsk(req)->rcv_nxt,
+			   tcp_rsk(req)->rcv_nxt +
+			   tcp_synack_window(req))) {
 		/* Out of window: send ACK and drop. */
 		if (!(flg & TCP_FLAG_RST) &&
 		    !tcp_oow_rate_limited(sock_net(sk), skb,
@@ -808,6 +810,8 @@ struct sock *tcp_check_req(struct sock *sk, struct sk_buff *skb,
 			req->rsk_ops->send_ack(sk, skb, req);
 		if (paws_reject)
 			NET_INC_STATS(sock_net(sk), LINUX_MIB_PAWSESTABREJECTED);
+		else if (tsecr_reject)
+			NET_INC_STATS(sock_net(sk), LINUX_MIB_TSECRREJECTED);
 		return NULL;
 	}
 
diff --git a/net/ipv4/tcp_output.c b/net/ipv4/tcp_output.c
index 9a3cf51eab78..0a660075add5 100644
--- a/net/ipv4/tcp_output.c
+++ b/net/ipv4/tcp_output.c
@@ -943,6 +943,12 @@ static unsigned int tcp_synack_options(const struct sock *sk,
 		opts->options |= OPTION_TS;
 		opts->tsval = tcp_skb_timestamp_ts(tcp_rsk(req)->req_usec_ts, skb) +
 			      tcp_rsk(req)->ts_off;
+		if (!tcp_rsk(req)->snt_tsval_first) {
+			if (!opts->tsval)
+				opts->tsval = ~0U;
+			tcp_rsk(req)->snt_tsval_first = opts->tsval;
+		}
+		WRITE_ONCE(tcp_rsk(req)->snt_tsval_last, opts->tsval);
 		opts->tsecr = READ_ONCE(req->ts_recent);
 		remaining -= TCPOLEN_TSTAMP_ALIGNED;
 	}
-- 
2.51.0


From c1d6d629ab0b3dd991c7fb04587dd2adf2c4b426 Mon Sep 17 00:00:00 2001
From: Willem de Bruijn <willemb@google.com>
Date: Mon, 24 Feb 2025 21:23:58 -0500
Subject: [PATCH 06/16] selftests/net: prepare cmsg_ipv6.sh for ipv4

Move IPV6_TCLASS and IPV6_HOPLIMIT into loops, to be able to use them
for IP_TOS and IP_TTL in a follow-on patch.

Indentation in this file is a mix of four spaces and tabs for double
indents. To minimize code churn, maintain that pattern.

Very small diff if viewing with -w.

Signed-off-by: Willem de Bruijn <willemb@google.com>
Link: https://patch.msgid.link/20250225022431.2083926-2-willemdebruijn.kernel@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 tools/testing/selftests/net/cmsg_ipv6.sh | 109 ++++++++++++-----------
 1 file changed, 59 insertions(+), 50 deletions(-)

diff --git a/tools/testing/selftests/net/cmsg_ipv6.sh b/tools/testing/selftests/net/cmsg_ipv6.sh
index 8bc23fb4c82b..51132c26e9b5 100755
--- a/tools/testing/selftests/net/cmsg_ipv6.sh
+++ b/tools/testing/selftests/net/cmsg_ipv6.sh
@@ -64,76 +64,85 @@ for ovr in setsock cmsg both diff; do
 done
 
 # IPV6_TCLASS
-TOS=0x10
-TOS2=0x20
 
-ip -6 -netns $NS rule add tos $TOS lookup 300
-ip -6 -netns $NS route add table 300 prohibit any
+test_dscp() {
+    local -r TOS=0x10
+    local -r TOS2=0x20
 
-for ovr in setsock cmsg both diff; do
-    for p in u i r; do
-	[ $p == "u" ] && prot=UDP
-	[ $p == "i" ] && prot=ICMP
-	[ $p == "r" ] && prot=RAW
+    ip -6 -netns $NS rule add tos $TOS lookup 300
+    ip -6 -netns $NS route add table 300 prohibit any
 
-	[ $ovr == "setsock" ] && m="-C"
-	[ $ovr == "cmsg" ]    && m="-c"
-	[ $ovr == "both" ]    && m="-C $((TOS2)) -c"
-	[ $ovr == "diff" ]    && m="-C $((TOS )) -c"
+    for ovr in setsock cmsg both diff; do
+	for p in u i r; do
+	    [ $p == "u" ] && prot=UDP
+	    [ $p == "i" ] && prot=ICMP
+	    [ $p == "r" ] && prot=RAW
 
-	$NSEXE nohup tcpdump --immediate-mode -p -ni dummy0 -w $TMPF -c 4 2> /dev/null &
-	BG=$!
-	sleep 0.05
+	    [ $ovr == "setsock" ] && m="-C"
+	    [ $ovr == "cmsg" ]    && m="-c"
+	    [ $ovr == "both" ]    && m="-C $((TOS2)) -c"
+	    [ $ovr == "diff" ]    && m="-C $((TOS )) -c"
 
-	$NSEXE ./cmsg_sender -6 -p $p $m $((TOS2)) $TGT6 1234
-	check_result $? 0 "TCLASS $prot $ovr - pass"
+	    $NSEXE nohup tcpdump --immediate-mode -p -ni dummy0 -w $TMPF -c 4 2> /dev/null &
+	    BG=$!
+	    sleep 0.05
 
-	while [ -d /proc/$BG ]; do
 	    $NSEXE ./cmsg_sender -6 -p $p $m $((TOS2)) $TGT6 1234
-	done
+	    check_result $? 0 "TCLASS $prot $ovr - pass"
 
-	tcpdump -r $TMPF -v 2>&1 | grep "class $TOS2" >> /dev/null
-	check_result $? 0 "TCLASS $prot $ovr - packet data"
-	rm $TMPF
+	    while [ -d /proc/$BG ]; do
+	        $NSEXE ./cmsg_sender -6 -p $p $m $((TOS2)) $TGT6 1234
+	    done
 
-	[ $ovr == "both" ]    && m="-C $((TOS )) -c"
-	[ $ovr == "diff" ]    && m="-C $((TOS2)) -c"
+	    tcpdump -r $TMPF -v 2>&1 | grep "class $TOS2" >> /dev/null
+	    check_result $? 0 "TCLASS $prot $ovr - packet data"
+	    rm $TMPF
 
-	$NSEXE ./cmsg_sender -6 -p $p $m $((TOS)) -s $TGT6 1234
-	check_result $? 1 "TCLASS $prot $ovr - rejection"
+	    [ $ovr == "both" ]    && m="-C $((TOS )) -c"
+	    [ $ovr == "diff" ]    && m="-C $((TOS2)) -c"
+
+	    $NSEXE ./cmsg_sender -6 -p $p $m $((TOS)) -s $TGT6 1234
+	    check_result $? 1 "TCLASS $prot $ovr - rejection"
+	done
     done
-done
+}
 
-# IPV6_HOPLIMIT
-LIM=4
+test_dscp
 
-for ovr in setsock cmsg both diff; do
-    for p in u i r; do
-	[ $p == "u" ] && prot=UDP
-	[ $p == "i" ] && prot=ICMP
-	[ $p == "r" ] && prot=RAW
+# IPV6_HOPLIMIT
+test_hoplimit() {
+    local -r LIM=4
 
-	[ $ovr == "setsock" ] && m="-L"
-	[ $ovr == "cmsg" ]    && m="-l"
-	[ $ovr == "both" ]    && m="-L $LIM -l"
-	[ $ovr == "diff" ]    && m="-L $((LIM + 1)) -l"
+    for ovr in setsock cmsg both diff; do
+	for p in u i r; do
+	    [ $p == "u" ] && prot=UDP
+	    [ $p == "i" ] && prot=ICMP
+	    [ $p == "r" ] && prot=RAW
 
-	$NSEXE nohup tcpdump --immediate-mode -p -ni dummy0 -w $TMPF -c 4 2> /dev/null &
-	BG=$!
-	sleep 0.05
+	    [ $ovr == "setsock" ] && m="-L"
+	    [ $ovr == "cmsg" ]    && m="-l"
+	    [ $ovr == "both" ]    && m="-L $LIM -l"
+	    [ $ovr == "diff" ]    && m="-L $((LIM + 1)) -l"
 
-	$NSEXE ./cmsg_sender -6 -p $p $m $LIM $TGT6 1234
-	check_result $? 0 "HOPLIMIT $prot $ovr - pass"
+	    $NSEXE nohup tcpdump --immediate-mode -p -ni dummy0 -w $TMPF -c 4 2> /dev/null &
+	    BG=$!
+	    sleep 0.05
 
-	while [ -d /proc/$BG ]; do
 	    $NSEXE ./cmsg_sender -6 -p $p $m $LIM $TGT6 1234
-	done
+	    check_result $? 0 "HOPLIMIT $prot $ovr - pass"
 
-	tcpdump -r $TMPF -v 2>&1 | grep "hlim $LIM[^0-9]" >> /dev/null
-	check_result $? 0 "HOPLIMIT $prot $ovr - packet data"
-	rm $TMPF
+	    while [ -d /proc/$BG ]; do
+	        $NSEXE ./cmsg_sender -6 -p $p $m $LIM $TGT6 1234
+	    done
+
+	    tcpdump -r $TMPF -v 2>&1 | grep "hlim $LIM[^0-9]" >> /dev/null
+	    check_result $? 0 "HOPLIMIT $prot $ovr - packet data"
+	    rm $TMPF
+	done
     done
-done
+}
+
+test_hoplimit
 
 # IPV6 exthdr
 for p in u i r; do
-- 
2.51.0


From 2e5584e0f913b53630238223443ae2a67eea714c Mon Sep 17 00:00:00 2001
From: Willem de Bruijn <willemb@google.com>
Date: Mon, 24 Feb 2025 21:23:59 -0500
Subject: [PATCH 07/16] selftests/net: expand cmsg_ipv6.sh with ipv4

Expand IPV6_TCLASS to also cover IP_TOS.
Expand IPV6_HOPLIMIT to also cover IP_TTL.

Expand csmg_sender.c to allow setting IPv4 setsockopts.
Also rename struct v6 to cmsg to match its expanded scope.
Don't bother updating all occurrences of tclass and hoplimit.

Rename cmsg_ipv6.sh to cmsg_ip.sh to match the expanded scope.

Be careful around the subtle API difference between TCLASS and TOS.
IP_TOS includes ECN bits. Add a test to verify that these are masked
when making routing decisions.

Diff is more concise with --word-diff

Signed-off-by: Willem de Bruijn <willemb@google.com>
Link: https://patch.msgid.link/20250225022431.2083926-3-willemdebruijn.kernel@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 tools/testing/selftests/net/Makefile          |  2 +-
 .../net/{cmsg_ipv6.sh => cmsg_ip.sh}          | 59 ++++++++----
 tools/testing/selftests/net/cmsg_sender.c     | 90 +++++++++++--------
 3 files changed, 96 insertions(+), 55 deletions(-)
 rename tools/testing/selftests/net/{cmsg_ipv6.sh => cmsg_ip.sh} (65%)

diff --git a/tools/testing/selftests/net/Makefile b/tools/testing/selftests/net/Makefile
index aeb96c085614..2e16ce6d28e4 100644
--- a/tools/testing/selftests/net/Makefile
+++ b/tools/testing/selftests/net/Makefile
@@ -34,7 +34,7 @@ TEST_PROGS += gre_gso.sh
 TEST_PROGS += cmsg_so_mark.sh
 TEST_PROGS += cmsg_so_priority.sh
 TEST_PROGS += test_so_rcv.sh
-TEST_PROGS += cmsg_time.sh cmsg_ipv6.sh
+TEST_PROGS += cmsg_time.sh cmsg_ip.sh
 TEST_PROGS += netns-name.sh
 TEST_PROGS += link_netns.py
 TEST_PROGS += nl_netdev.py
diff --git a/tools/testing/selftests/net/cmsg_ipv6.sh b/tools/testing/selftests/net/cmsg_ip.sh
similarity index 65%
rename from tools/testing/selftests/net/cmsg_ipv6.sh
rename to tools/testing/selftests/net/cmsg_ip.sh
index 51132c26e9b5..2a52520aca32 100755
--- a/tools/testing/selftests/net/cmsg_ipv6.sh
+++ b/tools/testing/selftests/net/cmsg_ip.sh
@@ -3,6 +3,8 @@
 
 source lib.sh
 
+IP4=172.16.0.1/24
+TGT4=172.16.0.2
 IP6=2001:db8:1::1/64
 TGT6=2001:db8:1::2
 TMPF=$(mktemp --suffix ".pcap")
@@ -30,6 +32,7 @@ $NSEXE sysctl -w net.ipv4.ping_group_range='0 2147483647' > /dev/null
 # Connectivity
 ip -netns $NS link add type dummy
 ip -netns $NS link set dev dummy0 up
+ip -netns $NS addr add $IP4 dev dummy0
 ip -netns $NS addr add $IP6 dev dummy0
 
 # Test
@@ -63,14 +66,19 @@ for ovr in setsock cmsg both diff; do
     done
 done
 
-# IPV6_TCLASS
+# IP_TOS + IPV6_TCLASS
 
 test_dscp() {
+    local -r IPVER=$1
+    local -r TGT=$2
+    local -r MATCH=$3
+
     local -r TOS=0x10
     local -r TOS2=0x20
+    local -r ECN=0x3
 
-    ip -6 -netns $NS rule add tos $TOS lookup 300
-    ip -6 -netns $NS route add table 300 prohibit any
+    ip $IPVER -netns $NS rule add tos $TOS lookup 300
+    ip $IPVER -netns $NS route add table 300 prohibit any
 
     for ovr in setsock cmsg both diff; do
 	for p in u i r; do
@@ -87,30 +95,42 @@ test_dscp() {
 	    BG=$!
 	    sleep 0.05
 
-	    $NSEXE ./cmsg_sender -6 -p $p $m $((TOS2)) $TGT6 1234
-	    check_result $? 0 "TCLASS $prot $ovr - pass"
+	    $NSEXE ./cmsg_sender $IPVER -p $p $m $((TOS2)) $TGT 1234
+	    check_result $? 0 "$MATCH $prot $ovr - pass"
 
 	    while [ -d /proc/$BG ]; do
-	        $NSEXE ./cmsg_sender -6 -p $p $m $((TOS2)) $TGT6 1234
+	        $NSEXE ./cmsg_sender $IPVER -p $p $m $((TOS2)) $TGT 1234
 	    done
 
-	    tcpdump -r $TMPF -v 2>&1 | grep "class $TOS2" >> /dev/null
-	    check_result $? 0 "TCLASS $prot $ovr - packet data"
+	    tcpdump -r $TMPF -v 2>&1 | grep "$MATCH $TOS2" >> /dev/null
+	    check_result $? 0 "$MATCH $prot $ovr - packet data"
 	    rm $TMPF
 
 	    [ $ovr == "both" ]    && m="-C $((TOS )) -c"
 	    [ $ovr == "diff" ]    && m="-C $((TOS2)) -c"
 
-	    $NSEXE ./cmsg_sender -6 -p $p $m $((TOS)) -s $TGT6 1234
-	    check_result $? 1 "TCLASS $prot $ovr - rejection"
+	    # Match prohibit rule: expect failure
+	    $NSEXE ./cmsg_sender $IPVER -p $p $m $((TOS)) -s $TGT 1234
+	    check_result $? 1 "$MATCH $prot $ovr - rejection"
+
+	    # Match prohibit rule: IPv4 masks ECN: expect failure
+	    if [[ "$IPVER" == "-4" ]]; then
+		$NSEXE ./cmsg_sender $IPVER -p $p $m "$((TOS | ECN))" -s $TGT 1234
+		check_result $? 1 "$MATCH $prot $ovr - rejection (ECN)"
+	    fi
 	done
     done
 }
 
-test_dscp
+test_dscp -4 $TGT4 tos
+test_dscp -6 $TGT6 class
+
+# IP_TTL + IPV6_HOPLIMIT
+test_ttl_hoplimit() {
+    local -r IPVER=$1
+    local -r TGT=$2
+    local -r MATCH=$3
 
-# IPV6_HOPLIMIT
-test_hoplimit() {
     local -r LIM=4
 
     for ovr in setsock cmsg both diff; do
@@ -128,21 +148,22 @@ test_hoplimit() {
 	    BG=$!
 	    sleep 0.05
 
-	    $NSEXE ./cmsg_sender -6 -p $p $m $LIM $TGT6 1234
-	    check_result $? 0 "HOPLIMIT $prot $ovr - pass"
+	    $NSEXE ./cmsg_sender $IPVER -p $p $m $LIM $TGT 1234
+	    check_result $? 0 "$MATCH $prot $ovr - pass"
 
 	    while [ -d /proc/$BG ]; do
-	        $NSEXE ./cmsg_sender -6 -p $p $m $LIM $TGT6 1234
+		$NSEXE ./cmsg_sender $IPVER -p $p $m $LIM $TGT 1234
 	    done
 
-	    tcpdump -r $TMPF -v 2>&1 | grep "hlim $LIM[^0-9]" >> /dev/null
-	    check_result $? 0 "HOPLIMIT $prot $ovr - packet data"
+	    tcpdump -r $TMPF -v 2>&1 | grep "$MATCH $LIM[^0-9]" >> /dev/null
+	    check_result $? 0 "$MATCH $prot $ovr - packet data"
 	    rm $TMPF
 	done
     done
 }
 
-test_hoplimit
+test_ttl_hoplimit -4 $TGT4 ttl
+test_ttl_hoplimit -6 $TGT6 hlim
 
 # IPV6 exthdr
 for p in u i r; do
diff --git a/tools/testing/selftests/net/cmsg_sender.c b/tools/testing/selftests/net/cmsg_sender.c
index bc314382e4e1..19bd8499031b 100644
--- a/tools/testing/selftests/net/cmsg_sender.c
+++ b/tools/testing/selftests/net/cmsg_sender.c
@@ -72,7 +72,7 @@ struct options {
 		struct option_cmsg_u32 tclass;
 		struct option_cmsg_u32 hlimit;
 		struct option_cmsg_u32 exthdr;
-	} v6;
+	} cmsg;
 } opt = {
 	.size = 13,
 	.num_pkt = 1,
@@ -104,10 +104,10 @@ static void __attribute__((noreturn)) cs_usage(const char *bin)
 	       "\t\t-t      Enable time stamp reporting\n"
 	       "\t\t-f val  Set don't fragment via cmsg\n"
 	       "\t\t-F val  Set don't fragment via setsockopt\n"
-	       "\t\t-c val  Set TCLASS via cmsg\n"
-	       "\t\t-C val  Set TCLASS via setsockopt\n"
-	       "\t\t-l val  Set HOPLIMIT via cmsg\n"
-	       "\t\t-L val  Set HOPLIMIT via setsockopt\n"
+	       "\t\t-c val  Set TOS/TCLASS via cmsg\n"
+	       "\t\t-C val  Set TOS/TCLASS via setsockopt\n"
+	       "\t\t-l val  Set TTL/HOPLIMIT via cmsg\n"
+	       "\t\t-L val  Set TTL/HOPLIMIT via setsockopt\n"
 	       "\t\t-H type Add an IPv6 header option\n"
 	       "\t\t        (h = HOP; d = DST; r = RTDST)"
 	       "");
@@ -169,37 +169,37 @@ static void cs_parse_args(int argc, char *argv[])
 			opt.ts.ena = true;
 			break;
 		case 'f':
-			opt.v6.dontfrag.ena = true;
-			opt.v6.dontfrag.val = atoi(optarg);
+			opt.cmsg.dontfrag.ena = true;
+			opt.cmsg.dontfrag.val = atoi(optarg);
 			break;
 		case 'F':
 			opt.sockopt.dontfrag = atoi(optarg);
 			break;
 		case 'c':
-			opt.v6.tclass.ena = true;
-			opt.v6.tclass.val = atoi(optarg);
+			opt.cmsg.tclass.ena = true;
+			opt.cmsg.tclass.val = atoi(optarg);
 			break;
 		case 'C':
 			opt.sockopt.tclass = atoi(optarg);
 			break;
 		case 'l':
-			opt.v6.hlimit.ena = true;
-			opt.v6.hlimit.val = atoi(optarg);
+			opt.cmsg.hlimit.ena = true;
+			opt.cmsg.hlimit.val = atoi(optarg);
 			break;
 		case 'L':
 			opt.sockopt.hlimit = atoi(optarg);
 			break;
 		case 'H':
-			opt.v6.exthdr.ena = true;
+			opt.cmsg.exthdr.ena = true;
 			switch (optarg[0]) {
 			case 'h':
-				opt.v6.exthdr.val = IPV6_HOPOPTS;
+				opt.cmsg.exthdr.val = IPV6_HOPOPTS;
 				break;
 			case 'd':
-				opt.v6.exthdr.val = IPV6_DSTOPTS;
+				opt.cmsg.exthdr.val = IPV6_DSTOPTS;
 				break;
 			case 'r':
-				opt.v6.exthdr.val = IPV6_RTHDRDSTOPTS;
+				opt.cmsg.exthdr.val = IPV6_RTHDRDSTOPTS;
 				break;
 			default:
 				printf("Error: hdr type: %s\n", optarg);
@@ -261,12 +261,20 @@ cs_write_cmsg(int fd, struct msghdr *msg, char *cbuf, size_t cbuf_sz)
 			  SOL_SOCKET, SO_MARK, &opt.mark);
 	ca_write_cmsg_u32(cbuf, cbuf_sz, &cmsg_len,
 			  SOL_SOCKET, SO_PRIORITY, &opt.priority);
-	ca_write_cmsg_u32(cbuf, cbuf_sz, &cmsg_len,
-			  SOL_IPV6, IPV6_DONTFRAG, &opt.v6.dontfrag);
-	ca_write_cmsg_u32(cbuf, cbuf_sz, &cmsg_len,
-			  SOL_IPV6, IPV6_TCLASS, &opt.v6.tclass);
-	ca_write_cmsg_u32(cbuf, cbuf_sz, &cmsg_len,
-			  SOL_IPV6, IPV6_HOPLIMIT, &opt.v6.hlimit);
+
+	if (opt.sock.family == AF_INET) {
+		ca_write_cmsg_u32(cbuf, cbuf_sz, &cmsg_len,
+				  SOL_IP, IP_TOS, &opt.cmsg.tclass);
+		ca_write_cmsg_u32(cbuf, cbuf_sz, &cmsg_len,
+				  SOL_IP, IP_TTL, &opt.cmsg.hlimit);
+	} else {
+		ca_write_cmsg_u32(cbuf, cbuf_sz, &cmsg_len,
+				  SOL_IPV6, IPV6_DONTFRAG, &opt.cmsg.dontfrag);
+		ca_write_cmsg_u32(cbuf, cbuf_sz, &cmsg_len,
+				  SOL_IPV6, IPV6_TCLASS, &opt.cmsg.tclass);
+		ca_write_cmsg_u32(cbuf, cbuf_sz, &cmsg_len,
+				  SOL_IPV6, IPV6_HOPLIMIT, &opt.cmsg.hlimit);
+	}
 
 	if (opt.txtime.ena) {
 		__u64 txtime;
@@ -297,14 +305,14 @@ cs_write_cmsg(int fd, struct msghdr *msg, char *cbuf, size_t cbuf_sz)
 		*(__u32 *)CMSG_DATA(cmsg) = SOF_TIMESTAMPING_TX_SCHED |
 					    SOF_TIMESTAMPING_TX_SOFTWARE;
 	}
-	if (opt.v6.exthdr.ena) {
+	if (opt.cmsg.exthdr.ena) {
 		cmsg = (struct cmsghdr *)(cbuf + cmsg_len);
 		cmsg_len += CMSG_SPACE(8);
 		if (cbuf_sz < cmsg_len)
 			error(ERN_CMSG_WR, EFAULT, "cmsg buffer too small");
 
 		cmsg->cmsg_level = SOL_IPV6;
-		cmsg->cmsg_type = opt.v6.exthdr.val;
+		cmsg->cmsg_type = opt.cmsg.exthdr.val;
 		cmsg->cmsg_len = CMSG_LEN(8);
 		*(__u64 *)CMSG_DATA(cmsg) = 0;
 	}
@@ -405,23 +413,35 @@ static void ca_set_sockopts(int fd)
 	    setsockopt(fd, SOL_SOCKET, SO_MARK,
 		       &opt.sockopt.mark, sizeof(opt.sockopt.mark)))
 		error(ERN_SOCKOPT, errno, "setsockopt SO_MARK");
-	if (opt.sockopt.dontfrag &&
-	    setsockopt(fd, SOL_IPV6, IPV6_DONTFRAG,
-		       &opt.sockopt.dontfrag, sizeof(opt.sockopt.dontfrag)))
-		error(ERN_SOCKOPT, errno, "setsockopt IPV6_DONTFRAG");
-	if (opt.sockopt.tclass &&
-	    setsockopt(fd, SOL_IPV6, IPV6_TCLASS,
-		       &opt.sockopt.tclass, sizeof(opt.sockopt.tclass)))
-		error(ERN_SOCKOPT, errno, "setsockopt IPV6_TCLASS");
-	if (opt.sockopt.hlimit &&
-	    setsockopt(fd, SOL_IPV6, IPV6_UNICAST_HOPS,
-		       &opt.sockopt.hlimit, sizeof(opt.sockopt.hlimit)))
-		error(ERN_SOCKOPT, errno, "setsockopt IPV6_HOPLIMIT");
 	if (opt.sockopt.priority &&
 	    setsockopt(fd, SOL_SOCKET, SO_PRIORITY,
 		       &opt.sockopt.priority, sizeof(opt.sockopt.priority)))
 		error(ERN_SOCKOPT, errno, "setsockopt SO_PRIORITY");
 
+	if (opt.sock.family == AF_INET) {
+		if (opt.sockopt.tclass &&
+		    setsockopt(fd, SOL_IP, IP_TOS,
+			       &opt.sockopt.tclass, sizeof(opt.sockopt.tclass)))
+			error(ERN_SOCKOPT, errno, "setsockopt IP_TOS");
+		if (opt.sockopt.hlimit &&
+		    setsockopt(fd, SOL_IP, IP_TTL,
+			       &opt.sockopt.hlimit, sizeof(opt.sockopt.hlimit)))
+			error(ERN_SOCKOPT, errno, "setsockopt IP_TTL");
+	} else {
+		if (opt.sockopt.dontfrag &&
+		    setsockopt(fd, SOL_IPV6, IPV6_DONTFRAG,
+			       &opt.sockopt.dontfrag, sizeof(opt.sockopt.dontfrag)))
+			error(ERN_SOCKOPT, errno, "setsockopt IPV6_DONTFRAG");
+		if (opt.sockopt.tclass &&
+		    setsockopt(fd, SOL_IPV6, IPV6_TCLASS,
+			       &opt.sockopt.tclass, sizeof(opt.sockopt.tclass)))
+			error(ERN_SOCKOPT, errno, "setsockopt IPV6_TCLASS");
+		if (opt.sockopt.hlimit &&
+		    setsockopt(fd, SOL_IPV6, IPV6_UNICAST_HOPS,
+			       &opt.sockopt.hlimit, sizeof(opt.sockopt.hlimit)))
+			error(ERN_SOCKOPT, errno, "setsockopt IPV6_HOPLIMIT");
+	}
+
 	if (opt.txtime.ena) {
 		struct sock_txtime so_txtime = {
 			.clockid = CLOCK_MONOTONIC,
-- 
2.51.0


From e6116fc605574bb58c2016938ff24a7fbafe6e2a Mon Sep 17 00:00:00 2001
From: Willem de Bruijn <willemb@google.com>
Date: Mon, 24 Feb 2025 21:33:55 -0500
Subject: [PATCH 08/16] net: skb: free up one bit in tx_flags

The linked series wants to add skb tx completion timestamps.
That needs a bit in skb_shared_info.tx_flags, but all are in use.

A per-skb bit is only needed for features that are configured on a
per packet basis. Per socket features can be read from sk->sk_tsflags.

Per packet tsflags can be set in sendmsg using cmsg, but only those in
SOF_TIMESTAMPING_TX_RECORD_MASK.

Per packet tsflags can also be set without cmsg by sandwiching a
send inbetween two setsockopts:

    val |= SOF_TIMESTAMPING_$FEATURE;
    setsockopt(fd, SOL_SOCKET, SO_TIMESTAMPING, &val, sizeof(val));
    write(fd, buf, sz);
    val &= ~SOF_TIMESTAMPING_$FEATURE;
    setsockopt(fd, SOL_SOCKET, SO_TIMESTAMPING, &val, sizeof(val));

Changing a datapath test from skb_shinfo(skb)->tx_flags to
skb->sk->sk_tsflags can change behavior in that case, as the tx_flags
is written before the second setsockopt updates sk_tsflags.

Therefore, only bits can be reclaimed that cannot be set by cmsg and
are also highly unlikely to be used to target individual packets
otherwise.

Free up the bit currently used for SKBTX_HW_TSTAMP_USE_CYCLES. This
selects between clock and free running counter source for HW TX
timestamps. It is probable that all packets of the same socket will
always use the same source.

Link: https://lore.kernel.org/netdev/cover.1739988644.git.pav@iki.fi/
Signed-off-by: Willem de Bruijn <willemb@google.com>
Reviewed-by: Jason Xing <kerneljasonxing@gmail.com>
Reviewed-by: Gerhard Engleder <gerhard@engleder-embedded.com>
Link: https://patch.msgid.link/20250225023416.2088705-1-willemdebruijn.kernel@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/engleder/tsnep_main.c |  4 ++--
 drivers/net/ethernet/intel/igc/igc_main.c  |  3 ++-
 include/linux/skbuff.h                     |  5 ++---
 net/socket.c                               | 11 +----------
 4 files changed, 7 insertions(+), 16 deletions(-)

diff --git a/drivers/net/ethernet/engleder/tsnep_main.c b/drivers/net/ethernet/engleder/tsnep_main.c
index 0d030cb0b21c..3de4cb06e266 100644
--- a/drivers/net/ethernet/engleder/tsnep_main.c
+++ b/drivers/net/ethernet/engleder/tsnep_main.c
@@ -852,8 +852,8 @@ static bool tsnep_tx_poll(struct tsnep_tx *tx, int napi_budget)
 			struct skb_shared_hwtstamps hwtstamps;
 			u64 timestamp;
 
-			if (skb_shinfo(entry->skb)->tx_flags &
-			    SKBTX_HW_TSTAMP_USE_CYCLES)
+			if (entry->skb->sk &&
+			    READ_ONCE(entry->skb->sk->sk_tsflags) & SOF_TIMESTAMPING_BIND_PHC)
 				timestamp =
 					__le64_to_cpu(entry->desc_wb->counter);
 			else
diff --git a/drivers/net/ethernet/intel/igc/igc_main.c b/drivers/net/ethernet/intel/igc/igc_main.c
index 3044392e8ded..472f009630c9 100644
--- a/drivers/net/ethernet/intel/igc/igc_main.c
+++ b/drivers/net/ethernet/intel/igc/igc_main.c
@@ -1668,7 +1668,8 @@ done:
 		if (igc_request_tx_tstamp(adapter, skb, &tstamp_flags)) {
 			skb_shinfo(skb)->tx_flags |= SKBTX_IN_PROGRESS;
 			tx_flags |= IGC_TX_FLAGS_TSTAMP | tstamp_flags;
-			if (skb_shinfo(skb)->tx_flags & SKBTX_HW_TSTAMP_USE_CYCLES)
+			if (skb->sk &&
+			    READ_ONCE(skb->sk->sk_tsflags) & SOF_TIMESTAMPING_BIND_PHC)
 				tx_flags |= IGC_TX_FLAGS_TSTAMP_TIMER_1;
 		} else {
 			adapter->tx_hwtstamp_skipped++;
diff --git a/include/linux/skbuff.h b/include/linux/skbuff.h
index f2bb8473d99a..171aa15f6541 100644
--- a/include/linux/skbuff.h
+++ b/include/linux/skbuff.h
@@ -478,8 +478,8 @@ enum {
 	/* device driver is going to provide hardware time stamp */
 	SKBTX_IN_PROGRESS = 1 << 2,
 
-	/* generate hardware time stamp based on cycles if supported */
-	SKBTX_HW_TSTAMP_USE_CYCLES = 1 << 3,
+	/* reserved */
+	SKBTX_RESERVED = 1 << 3,
 
 	/* generate wifi status information (where possible) */
 	SKBTX_WIFI_STATUS = 1 << 4,
@@ -500,7 +500,6 @@ enum {
 				 SKBTX_SCHED_TSTAMP | \
 				 SKBTX_BPF)
 #define SKBTX_ANY_TSTAMP	(SKBTX_HW_TSTAMP | \
-				 SKBTX_HW_TSTAMP_USE_CYCLES | \
 				 SKBTX_ANY_SW_TSTAMP)
 
 /* Definitions for flags in struct skb_shared_info */
diff --git a/net/socket.c b/net/socket.c
index 0545e9ea7058..b64ecf2722e7 100644
--- a/net/socket.c
+++ b/net/socket.c
@@ -680,18 +680,9 @@ void __sock_tx_timestamp(__u32 tsflags, __u8 *tx_flags)
 {
 	u8 flags = *tx_flags;
 
-	if (tsflags & SOF_TIMESTAMPING_TX_HARDWARE) {
+	if (tsflags & SOF_TIMESTAMPING_TX_HARDWARE)
 		flags |= SKBTX_HW_TSTAMP_NOBPF;
 
-		/* PTP hardware clocks can provide a free running cycle counter
-		 * as a time base for virtual clocks. Tell driver to use the
-		 * free running cycle counter for timestamp if socket is bound
-		 * to virtual clock.
-		 */
-		if (tsflags & SOF_TIMESTAMPING_BIND_PHC)
-			flags |= SKBTX_HW_TSTAMP_USE_CYCLES;
-	}
-
 	if (tsflags & SOF_TIMESTAMPING_TX_SOFTWARE)
 		flags |= SKBTX_SW_TSTAMP;
 
-- 
2.51.0


From 28d68d396a1cd21591e8c6d74afbde33a7ea107e Mon Sep 17 00:00:00 2001
From: Hangbin Liu <liuhangbin@gmail.com>
Date: Tue, 25 Feb 2025 03:39:14 +0000
Subject: [PATCH 09/16] bonding: report duplicate MAC address in all situations
MIME-Version: 1.0
Content-Type: text/plain; charset=utf8
Content-Transfer-Encoding: 8bit

Normally, a bond uses the MAC address of the first added slave as the bond’s
MAC address. And the bond will set active slave’s MAC address to bond’s
address if fail_over_mac is set to none (0) or follow (2).

When the first slave is removed, the bond will still use the removed slave’s
MAC address, which can lead to a duplicate MAC address and potentially cause
issues with the switch. To avoid confusion, let's warn the user in all
situations, including when fail_over_mac is set to 2 or not in active-backup
mode.

Signed-off-by: Hangbin Liu <liuhangbin@gmail.com>
Reviewed-by: Nikolay Aleksandrov <razor@blackwall.org>
Link: https://patch.msgid.link/20250225033914.18617-1-liuhangbin@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/bonding/bond_main.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/bonding/bond_main.c b/drivers/net/bonding/bond_main.c
index 7d716e90a84c..7d98fee5a27f 100644
--- a/drivers/net/bonding/bond_main.c
+++ b/drivers/net/bonding/bond_main.c
@@ -2548,7 +2548,7 @@ static int __bond_release_one(struct net_device *bond_dev,
 
 	RCU_INIT_POINTER(bond->current_arp_slave, NULL);
 
-	if (!all && (!bond->params.fail_over_mac ||
+	if (!all && (bond->params.fail_over_mac != BOND_FOM_ACTIVE ||
 		     BOND_MODE(bond) != BOND_MODE_ACTIVEBACKUP)) {
 		if (ether_addr_equal_64bits(bond_dev->dev_addr, slave->perm_hwaddr) &&
 		    bond_has_slaves(bond))
-- 
2.51.0


From bd7c00605ee0cf0bf27764d5da0b948d8004229e Mon Sep 17 00:00:00 2001
From: Ahmed Zaki <ahmed.zaki@intel.com>
Date: Mon, 24 Feb 2025 16:22:22 -0700
Subject: [PATCH 10/16] net: move aRFS rmap management and CPU affinity to core

A common task for most drivers is to remember the user-set CPU affinity
to its IRQs. On each netdev reset, the driver should re-assign the user's
settings to the IRQs. Unify this task across all drivers by moving the CPU
affinity to napi->config.

However, to move the CPU affinity to core, we also need to move aRFS
rmap management since aRFS uses its own IRQ notifiers.

For the aRFS, add a new netdev flag "rx_cpu_rmap_auto". Drivers supporting
aRFS should set the flag via netif_enable_cpu_rmap() and core will allocate
and manage the aRFS rmaps. Freeing the rmap is also done by core when the
netdev is freed. For better IRQ affinity management, move the IRQ rmap
notifier inside the napi_struct and add new notify.notify and
notify.release functions: netif_irq_cpu_rmap_notify() and
netif_napi_affinity_release().

Now we have the aRFS rmap management in core, add CPU affinity mask to
napi_config. To delegate the CPU affinity management to the core, drivers
must:
 1 - set the new netdev flag "irq_affinity_auto":
                                       netif_enable_irq_affinity(netdev)
 2 - create the napi with persistent config:
                                       netif_napi_add_config()
 3 - bind an IRQ to the napi instance: netif_napi_set_irq()

the core will then make sure to use re-assign affinity to the napi's
IRQ.

The default IRQ mask is set to one cpu starting from the closest NUMA.

Signed-off-by: Ahmed Zaki <ahmed.zaki@intel.com>
Link: https://patch.msgid.link/20250224232228.990783-2-ahmed.zaki@intel.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 Documentation/networking/scaling.rst |   6 +-
 include/linux/cpu_rmap.h             |   1 +
 include/linux/netdevice.h            |  24 +++-
 lib/cpu_rmap.c                       |   2 +-
 net/core/dev.c                       | 169 +++++++++++++++++++++++++++
 5 files changed, 195 insertions(+), 7 deletions(-)

diff --git a/Documentation/networking/scaling.rst b/Documentation/networking/scaling.rst
index 6984700cec6a..99b6a61e5e31 100644
--- a/Documentation/networking/scaling.rst
+++ b/Documentation/networking/scaling.rst
@@ -434,8 +434,10 @@ rps_dev_flow_table. The stack consults a CPU to hardware queue map which
 is maintained by the NIC driver. This is an auto-generated reverse map of
 the IRQ affinity table shown by /proc/interrupts. Drivers can use
 functions in the cpu_rmap (“CPU affinity reverse map”) kernel library
-to populate the map. For each CPU, the corresponding queue in the map is
-set to be one whose processing CPU is closest in cache locality.
+to populate the map. Alternatively, drivers can delegate the cpu_rmap
+management to the Kernel by calling netif_enable_cpu_rmap(). For each CPU,
+the corresponding queue in the map is set to be one whose processing CPU is
+closest in cache locality.
 
 
 Accelerated RFS Configuration
diff --git a/include/linux/cpu_rmap.h b/include/linux/cpu_rmap.h
index 20b5729903d7..2fd7ba75362a 100644
--- a/include/linux/cpu_rmap.h
+++ b/include/linux/cpu_rmap.h
@@ -32,6 +32,7 @@ struct cpu_rmap {
 #define CPU_RMAP_DIST_INF 0xffff
 
 extern struct cpu_rmap *alloc_cpu_rmap(unsigned int size, gfp_t flags);
+extern void cpu_rmap_get(struct cpu_rmap *rmap);
 extern int cpu_rmap_put(struct cpu_rmap *rmap);
 
 extern int cpu_rmap_add(struct cpu_rmap *rmap, void *obj);
diff --git a/include/linux/netdevice.h b/include/linux/netdevice.h
index 9a387d456592..2094d3edda73 100644
--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -352,6 +352,7 @@ struct napi_config {
 	u64 gro_flush_timeout;
 	u64 irq_suspend_timeout;
 	u32 defer_hard_irqs;
+	cpumask_t affinity_mask;
 	unsigned int napi_id;
 };
 
@@ -394,6 +395,8 @@ struct napi_struct {
 	struct list_head	dev_list;
 	struct hlist_node	napi_hash_node;
 	int			irq;
+	struct irq_affinity_notify notify;
+	int			napi_rmap_idx;
 	int			index;
 	struct napi_config	*config;
 };
@@ -409,6 +412,7 @@ enum {
 	NAPI_STATE_PREFER_BUSY_POLL,	/* prefer busy-polling over softirq processing*/
 	NAPI_STATE_THREADED,		/* The poll is performed inside its own thread*/
 	NAPI_STATE_SCHED_THREADED,	/* Napi is currently scheduled in threaded mode */
+	NAPI_STATE_HAS_NOTIFIER,	/* Napi has an IRQ notifier */
 };
 
 enum {
@@ -422,6 +426,7 @@ enum {
 	NAPIF_STATE_PREFER_BUSY_POLL	= BIT(NAPI_STATE_PREFER_BUSY_POLL),
 	NAPIF_STATE_THREADED		= BIT(NAPI_STATE_THREADED),
 	NAPIF_STATE_SCHED_THREADED	= BIT(NAPI_STATE_SCHED_THREADED),
+	NAPIF_STATE_HAS_NOTIFIER	= BIT(NAPI_STATE_HAS_NOTIFIER),
 };
 
 enum gro_result {
@@ -1989,6 +1994,15 @@ enum netdev_reg_state {
  *
  *	@threaded:	napi threaded mode is enabled
  *
+ *	@irq_affinity_auto: driver wants the core to store and re-assign the IRQ
+ *			    affinity. Set by netif_enable_irq_affinity(), then
+ *			    the driver must create a persistent napi by
+ *			    netif_napi_add_config() and finally bind the napi to
+ *			    IRQ (via netif_napi_set_irq()).
+ *
+ *	@rx_cpu_rmap_auto: driver wants the core to manage the ARFS rmap.
+ *	                   Set by calling netif_enable_cpu_rmap().
+ *
  *	@see_all_hwtstamp_requests: device wants to see calls to
  *			ndo_hwtstamp_set() for all timestamp requests
  *			regardless of source, even if those aren't
@@ -2396,6 +2410,8 @@ struct net_device {
 	struct lock_class_key	*qdisc_tx_busylock;
 	bool			proto_down;
 	bool			threaded;
+	bool			irq_affinity_auto;
+	bool			rx_cpu_rmap_auto;
 
 	/* priv_flags_slow, ungrouped to save space */
 	unsigned long		see_all_hwtstamp_requests:1;
@@ -2724,10 +2740,7 @@ static inline void netdev_assert_locked_or_invisible(struct net_device *dev)
 		netdev_assert_locked(dev);
 }
 
-static inline void netif_napi_set_irq_locked(struct napi_struct *napi, int irq)
-{
-	napi->irq = irq;
-}
+void netif_napi_set_irq_locked(struct napi_struct *napi, int irq);
 
 static inline void netif_napi_set_irq(struct napi_struct *napi, int irq)
 {
@@ -2865,6 +2878,9 @@ static inline void netif_napi_del(struct napi_struct *napi)
 	synchronize_net();
 }
 
+int netif_enable_cpu_rmap(struct net_device *dev, unsigned int num_irqs);
+void netif_set_affinity_auto(struct net_device *dev);
+
 struct packet_type {
 	__be16			type;	/* This is really htons(ether_type). */
 	bool			ignore_outgoing;
diff --git a/lib/cpu_rmap.c b/lib/cpu_rmap.c
index 4c348670da31..f03d9be3f06b 100644
--- a/lib/cpu_rmap.c
+++ b/lib/cpu_rmap.c
@@ -73,7 +73,7 @@ static void cpu_rmap_release(struct kref *ref)
  * cpu_rmap_get - internal helper to get new ref on a cpu_rmap
  * @rmap: reverse-map allocated with alloc_cpu_rmap()
  */
-static inline void cpu_rmap_get(struct cpu_rmap *rmap)
+void cpu_rmap_get(struct cpu_rmap *rmap)
 {
 	kref_get(&rmap->refcount);
 }
diff --git a/net/core/dev.c b/net/core/dev.c
index 3f525278a871..b0ab0169f507 100644
--- a/net/core/dev.c
+++ b/net/core/dev.c
@@ -6943,11 +6943,175 @@ void netif_queue_set_napi(struct net_device *dev, unsigned int queue_index,
 }
 EXPORT_SYMBOL(netif_queue_set_napi);
 
+static void
+netif_napi_irq_notify(struct irq_affinity_notify *notify,
+		      const cpumask_t *mask)
+{
+	struct napi_struct *napi =
+		container_of(notify, struct napi_struct, notify);
+#ifdef CONFIG_RFS_ACCEL
+	struct cpu_rmap *rmap = napi->dev->rx_cpu_rmap;
+	int err;
+#endif
+
+	if (napi->config && napi->dev->irq_affinity_auto)
+		cpumask_copy(&napi->config->affinity_mask, mask);
+
+#ifdef CONFIG_RFS_ACCEL
+	if (napi->dev->rx_cpu_rmap_auto) {
+		err = cpu_rmap_update(rmap, napi->napi_rmap_idx, mask);
+		if (err)
+			netdev_warn(napi->dev, "RMAP update failed (%d)\n",
+				    err);
+	}
+#endif
+}
+
+#ifdef CONFIG_RFS_ACCEL
+static void netif_napi_affinity_release(struct kref *ref)
+{
+	struct napi_struct *napi =
+		container_of(ref, struct napi_struct, notify.kref);
+	struct cpu_rmap *rmap = napi->dev->rx_cpu_rmap;
+
+	netdev_assert_locked(napi->dev);
+	WARN_ON(test_and_clear_bit(NAPI_STATE_HAS_NOTIFIER,
+				   &napi->state));
+
+	if (!napi->dev->rx_cpu_rmap_auto)
+		return;
+	rmap->obj[napi->napi_rmap_idx] = NULL;
+	napi->napi_rmap_idx = -1;
+	cpu_rmap_put(rmap);
+}
+
+int netif_enable_cpu_rmap(struct net_device *dev, unsigned int num_irqs)
+{
+	if (dev->rx_cpu_rmap_auto)
+		return 0;
+
+	dev->rx_cpu_rmap = alloc_irq_cpu_rmap(num_irqs);
+	if (!dev->rx_cpu_rmap)
+		return -ENOMEM;
+
+	dev->rx_cpu_rmap_auto = true;
+	return 0;
+}
+EXPORT_SYMBOL(netif_enable_cpu_rmap);
+
+static void netif_del_cpu_rmap(struct net_device *dev)
+{
+	struct cpu_rmap *rmap = dev->rx_cpu_rmap;
+
+	if (!dev->rx_cpu_rmap_auto)
+		return;
+
+	/* Free the rmap */
+	cpu_rmap_put(rmap);
+	dev->rx_cpu_rmap = NULL;
+	dev->rx_cpu_rmap_auto = false;
+}
+
+#else
+static void netif_napi_affinity_release(struct kref *ref)
+{
+}
+
+int netif_enable_cpu_rmap(struct net_device *dev, unsigned int num_irqs)
+{
+	return 0;
+}
+EXPORT_SYMBOL(netif_enable_cpu_rmap);
+
+static void netif_del_cpu_rmap(struct net_device *dev)
+{
+}
+#endif
+
+void netif_set_affinity_auto(struct net_device *dev)
+{
+	unsigned int i, maxqs, numa;
+
+	maxqs = max(dev->num_tx_queues, dev->num_rx_queues);
+	numa = dev_to_node(&dev->dev);
+
+	for (i = 0; i < maxqs; i++)
+		cpumask_set_cpu(cpumask_local_spread(i, numa),
+				&dev->napi_config[i].affinity_mask);
+
+	dev->irq_affinity_auto = true;
+}
+EXPORT_SYMBOL(netif_set_affinity_auto);
+
+void netif_napi_set_irq_locked(struct napi_struct *napi, int irq)
+{
+	int rc;
+
+	netdev_assert_locked_or_invisible(napi->dev);
+
+	if (napi->irq == irq)
+		return;
+
+	/* Remove existing resources */
+	if (test_and_clear_bit(NAPI_STATE_HAS_NOTIFIER, &napi->state))
+		irq_set_affinity_notifier(napi->irq, NULL);
+
+	napi->irq = irq;
+	if (irq < 0 ||
+	    (!napi->dev->rx_cpu_rmap_auto && !napi->dev->irq_affinity_auto))
+		return;
+
+	/* Abort for buggy drivers */
+	if (napi->dev->irq_affinity_auto && WARN_ON_ONCE(!napi->config))
+		return;
+
+#ifdef CONFIG_RFS_ACCEL
+	if (napi->dev->rx_cpu_rmap_auto) {
+		rc = cpu_rmap_add(napi->dev->rx_cpu_rmap, napi);
+		if (rc < 0)
+			return;
+
+		cpu_rmap_get(napi->dev->rx_cpu_rmap);
+		napi->napi_rmap_idx = rc;
+	}
+#endif
+
+	/* Use core IRQ notifier */
+	napi->notify.notify = netif_napi_irq_notify;
+	napi->notify.release = netif_napi_affinity_release;
+	rc = irq_set_affinity_notifier(irq, &napi->notify);
+	if (rc) {
+		netdev_warn(napi->dev, "Unable to set IRQ notifier (%d)\n",
+			    rc);
+		goto put_rmap;
+	}
+
+	set_bit(NAPI_STATE_HAS_NOTIFIER, &napi->state);
+	return;
+
+put_rmap:
+#ifdef CONFIG_RFS_ACCEL
+	if (napi->dev->rx_cpu_rmap_auto) {
+		cpu_rmap_put(napi->dev->rx_cpu_rmap);
+		napi->dev->rx_cpu_rmap->obj[napi->napi_rmap_idx] = NULL;
+		napi->napi_rmap_idx = -1;
+	}
+#endif
+	napi->notify.notify = NULL;
+	napi->notify.release = NULL;
+}
+EXPORT_SYMBOL(netif_napi_set_irq_locked);
+
 static void napi_restore_config(struct napi_struct *n)
 {
 	n->defer_hard_irqs = n->config->defer_hard_irqs;
 	n->gro_flush_timeout = n->config->gro_flush_timeout;
 	n->irq_suspend_timeout = n->config->irq_suspend_timeout;
+
+	if (n->dev->irq_affinity_auto &&
+	    test_bit(NAPI_STATE_HAS_NOTIFIER, &n->state))
+		irq_set_affinity(n->irq, &n->config->affinity_mask);
+
 	/* a NAPI ID might be stored in the config, if so use it. if not, use
 	 * napi_hash_add to generate one for us.
 	 */
@@ -7168,6 +7332,9 @@ void __netif_napi_del_locked(struct napi_struct *napi)
 	/* Make sure NAPI is disabled (or was never enabled). */
 	WARN_ON(!test_bit(NAPI_STATE_SCHED, &napi->state));
 
+	if (test_and_clear_bit(NAPI_STATE_HAS_NOTIFIER, &napi->state))
+		irq_set_affinity_notifier(napi->irq, NULL);
+
 	if (napi->config) {
 		napi->index = -1;
 		napi->config = NULL;
@@ -11720,6 +11887,8 @@ void free_netdev(struct net_device *dev)
 
 	netdev_napi_exit(dev);
 
+	netif_del_cpu_rmap(dev);
+
 	ref_tracker_dir_exit(&dev->refcnt_tracker);
 #ifdef CONFIG_PCPU_DEV_REFCNT
 	free_percpu(dev->pcpu_refcnt);
-- 
2.51.0


From de340d8206bf1897fcc05436adffb4d63a0a13e0 Mon Sep 17 00:00:00 2001
From: Ahmed Zaki <ahmed.zaki@intel.com>
Date: Mon, 24 Feb 2025 16:22:23 -0700
Subject: [PATCH 11/16] net: ena: use napi's aRFS rmap notifers

Use the core's rmap notifiers and delete our own.

Acked-by: David Arinzon <darinzon@amazon.com>
Signed-off-by: Ahmed Zaki <ahmed.zaki@intel.com>
Link: https://patch.msgid.link/20250224232228.990783-3-ahmed.zaki@intel.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/amazon/ena/ena_netdev.c | 43 +-------------------
 1 file changed, 1 insertion(+), 42 deletions(-)

diff --git a/drivers/net/ethernet/amazon/ena/ena_netdev.c b/drivers/net/ethernet/amazon/ena/ena_netdev.c
index c1295dfad0d0..6aab85a7c60a 100644
--- a/drivers/net/ethernet/amazon/ena/ena_netdev.c
+++ b/drivers/net/ethernet/amazon/ena/ena_netdev.c
@@ -5,9 +5,6 @@
 
 #define pr_fmt(fmt) KBUILD_MODNAME ": " fmt
 
-#ifdef CONFIG_RFS_ACCEL
-#include <linux/cpu_rmap.h>
-#endif /* CONFIG_RFS_ACCEL */
 #include <linux/ethtool.h>
 #include <linux/kernel.h>
 #include <linux/module.h>
@@ -162,30 +159,6 @@ int ena_xmit_common(struct ena_adapter *adapter,
 	return 0;
 }
 
-static int ena_init_rx_cpu_rmap(struct ena_adapter *adapter)
-{
-#ifdef CONFIG_RFS_ACCEL
-	u32 i;
-	int rc;
-
-	adapter->netdev->rx_cpu_rmap = alloc_irq_cpu_rmap(adapter->num_io_queues);
-	if (!adapter->netdev->rx_cpu_rmap)
-		return -ENOMEM;
-	for (i = 0; i < adapter->num_io_queues; i++) {
-		int irq_idx = ENA_IO_IRQ_IDX(i);
-
-		rc = irq_cpu_rmap_add(adapter->netdev->rx_cpu_rmap,
-				      pci_irq_vector(adapter->pdev, irq_idx));
-		if (rc) {
-			free_irq_cpu_rmap(adapter->netdev->rx_cpu_rmap);
-			adapter->netdev->rx_cpu_rmap = NULL;
-			return rc;
-		}
-	}
-#endif /* CONFIG_RFS_ACCEL */
-	return 0;
-}
-
 static void ena_init_io_rings_common(struct ena_adapter *adapter,
 				     struct ena_ring *ring, u16 qid)
 {
@@ -1596,7 +1569,7 @@ static int ena_enable_msix(struct ena_adapter *adapter)
 		adapter->num_io_queues = irq_cnt - ENA_ADMIN_MSIX_VEC;
 	}
 
-	if (ena_init_rx_cpu_rmap(adapter))
+	if (netif_enable_cpu_rmap(adapter->netdev, adapter->num_io_queues))
 		netif_warn(adapter, probe, adapter->netdev,
 			   "Failed to map IRQs to CPUs\n");
 
@@ -1742,13 +1715,6 @@ static void ena_free_io_irq(struct ena_adapter *adapter)
 	struct ena_irq *irq;
 	int i;
 
-#ifdef CONFIG_RFS_ACCEL
-	if (adapter->msix_vecs >= 1) {
-		free_irq_cpu_rmap(adapter->netdev->rx_cpu_rmap);
-		adapter->netdev->rx_cpu_rmap = NULL;
-	}
-#endif /* CONFIG_RFS_ACCEL */
-
 	for (i = ENA_IO_IRQ_FIRST_IDX; i < ENA_MAX_MSIX_VEC(io_queue_count); i++) {
 		irq = &adapter->irq_tbl[i];
 		irq_set_affinity_hint(irq->vector, NULL);
@@ -4131,13 +4097,6 @@ static void __ena_shutoff(struct pci_dev *pdev, bool shutdown)
 	ena_dev = adapter->ena_dev;
 	netdev = adapter->netdev;
 
-#ifdef CONFIG_RFS_ACCEL
-	if ((adapter->msix_vecs >= 1) && (netdev->rx_cpu_rmap)) {
-		free_irq_cpu_rmap(netdev->rx_cpu_rmap);
-		netdev->rx_cpu_rmap = NULL;
-	}
-
-#endif /* CONFIG_RFS_ACCEL */
 	/* Make sure timer and reset routine won't be called after
 	 * freeing device resources.
 	 */
-- 
2.51.0


From 30b78ba3d4fe7a4dcf86f7ed21e719a62b71a392 Mon Sep 17 00:00:00 2001
From: Ahmed Zaki <ahmed.zaki@intel.com>
Date: Mon, 24 Feb 2025 16:22:24 -0700
Subject: [PATCH 12/16] ice: clear NAPI's IRQ numbers in
 ice_vsi_clear_napi_queues()

We set the NAPI's IRQ number in ice_vsi_set_napi_queues(). Clear the
NAPI's IRQ in ice_vsi_clear_napi_queues().

Signed-off-by: Ahmed Zaki <ahmed.zaki@intel.com>
Link: https://patch.msgid.link/20250224232228.990783-4-ahmed.zaki@intel.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/intel/ice/ice_lib.c | 9 ++++++++-
 1 file changed, 8 insertions(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/intel/ice/ice_lib.c b/drivers/net/ethernet/intel/ice/ice_lib.c
index 7f5b229cab05..ce30674abf8f 100644
--- a/drivers/net/ethernet/intel/ice/ice_lib.c
+++ b/drivers/net/ethernet/intel/ice/ice_lib.c
@@ -2765,11 +2765,18 @@ void ice_vsi_set_napi_queues(struct ice_vsi *vsi)
 void ice_vsi_clear_napi_queues(struct ice_vsi *vsi)
 {
 	struct net_device *netdev = vsi->netdev;
-	int q_idx;
+	int q_idx, v_idx;
 
 	if (!netdev)
 		return;
 
+	/* Clear the NAPI's interrupt number */
+	ice_for_each_q_vector(vsi, v_idx) {
+		struct ice_q_vector *q_vector = vsi->q_vectors[v_idx];
+
+		netif_napi_set_irq(&q_vector->napi, -1);
+	}
+
 	ice_for_each_txq(vsi, q_idx)
 		netif_queue_set_napi(netdev, q_idx, NETDEV_QUEUE_TYPE_TX, NULL);
 
-- 
2.51.0


From 4063af296762e239588e0f69abc268e82f651983 Mon Sep 17 00:00:00 2001
From: Ahmed Zaki <ahmed.zaki@intel.com>
Date: Mon, 24 Feb 2025 16:22:25 -0700
Subject: [PATCH 13/16] ice: use napi's irq affinity and rmap IRQ notifiers

Delete the driver CPU affinity and aRFS rmap info, use the core's
API instead.

Signed-off-by: Ahmed Zaki <ahmed.zaki@intel.com>
Link: https://patch.msgid.link/20250224232228.990783-5-ahmed.zaki@intel.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/intel/ice/ice.h      |  3 --
 drivers/net/ethernet/intel/ice/ice_arfs.c | 33 +---------------
 drivers/net/ethernet/intel/ice/ice_arfs.h |  2 -
 drivers/net/ethernet/intel/ice/ice_base.c |  7 +---
 drivers/net/ethernet/intel/ice/ice_lib.c  |  7 ----
 drivers/net/ethernet/intel/ice/ice_main.c | 47 ++---------------------
 6 files changed, 6 insertions(+), 93 deletions(-)

diff --git a/drivers/net/ethernet/intel/ice/ice.h b/drivers/net/ethernet/intel/ice/ice.h
index c9104b13e1d2..5ceeae664598 100644
--- a/drivers/net/ethernet/intel/ice/ice.h
+++ b/drivers/net/ethernet/intel/ice/ice.h
@@ -475,9 +475,6 @@ struct ice_q_vector {
 	struct ice_ring_container rx;
 	struct ice_ring_container tx;
 
-	cpumask_t affinity_mask;
-	struct irq_affinity_notify affinity_notify;
-
 	struct ice_channel *ch;
 
 	char name[ICE_INT_NAME_STR_LEN];
diff --git a/drivers/net/ethernet/intel/ice/ice_arfs.c b/drivers/net/ethernet/intel/ice/ice_arfs.c
index 7cee365cc7d1..171cdec741c2 100644
--- a/drivers/net/ethernet/intel/ice/ice_arfs.c
+++ b/drivers/net/ethernet/intel/ice/ice_arfs.c
@@ -570,25 +570,6 @@ void ice_clear_arfs(struct ice_vsi *vsi)
 	vsi->arfs_fltr_cntrs = NULL;
 }
 
-/**
- * ice_free_cpu_rx_rmap - free setup CPU reverse map
- * @vsi: the VSI to be forwarded to
- */
-void ice_free_cpu_rx_rmap(struct ice_vsi *vsi)
-{
-	struct net_device *netdev;
-
-	if (!vsi || vsi->type != ICE_VSI_PF)
-		return;
-
-	netdev = vsi->netdev;
-	if (!netdev || !netdev->rx_cpu_rmap)
-		return;
-
-	free_irq_cpu_rmap(netdev->rx_cpu_rmap);
-	netdev->rx_cpu_rmap = NULL;
-}
-
 /**
  * ice_set_cpu_rx_rmap - setup CPU reverse map for each queue
  * @vsi: the VSI to be forwarded to
@@ -597,7 +578,6 @@ int ice_set_cpu_rx_rmap(struct ice_vsi *vsi)
 {
 	struct net_device *netdev;
 	struct ice_pf *pf;
-	int i;
 
 	if (!vsi || vsi->type != ICE_VSI_PF)
 		return 0;
@@ -610,18 +590,7 @@ int ice_set_cpu_rx_rmap(struct ice_vsi *vsi)
 	netdev_dbg(netdev, "Setup CPU RMAP: vsi type 0x%x, ifname %s, q_vectors %d\n",
 		   vsi->type, netdev->name, vsi->num_q_vectors);
 
-	netdev->rx_cpu_rmap = alloc_irq_cpu_rmap(vsi->num_q_vectors);
-	if (unlikely(!netdev->rx_cpu_rmap))
-		return -EINVAL;
-
-	ice_for_each_q_vector(vsi, i)
-		if (irq_cpu_rmap_add(netdev->rx_cpu_rmap,
-				     vsi->q_vectors[i]->irq.virq)) {
-			ice_free_cpu_rx_rmap(vsi);
-			return -EINVAL;
-		}
-
-	return 0;
+	return netif_enable_cpu_rmap(netdev, vsi->num_q_vectors);
 }
 
 /**
diff --git a/drivers/net/ethernet/intel/ice/ice_arfs.h b/drivers/net/ethernet/intel/ice/ice_arfs.h
index 9669ad9bf7b5..9706293128c3 100644
--- a/drivers/net/ethernet/intel/ice/ice_arfs.h
+++ b/drivers/net/ethernet/intel/ice/ice_arfs.h
@@ -45,7 +45,6 @@ int
 ice_rx_flow_steer(struct net_device *netdev, const struct sk_buff *skb,
 		  u16 rxq_idx, u32 flow_id);
 void ice_clear_arfs(struct ice_vsi *vsi);
-void ice_free_cpu_rx_rmap(struct ice_vsi *vsi);
 void ice_init_arfs(struct ice_vsi *vsi);
 void ice_sync_arfs_fltrs(struct ice_pf *pf);
 int ice_set_cpu_rx_rmap(struct ice_vsi *vsi);
@@ -56,7 +55,6 @@ ice_is_arfs_using_perfect_flow(struct ice_hw *hw,
 			       enum ice_fltr_ptype flow_type);
 #else
 static inline void ice_clear_arfs(struct ice_vsi *vsi) { }
-static inline void ice_free_cpu_rx_rmap(struct ice_vsi *vsi) { }
 static inline void ice_init_arfs(struct ice_vsi *vsi) { }
 static inline void ice_sync_arfs_fltrs(struct ice_pf *pf) { }
 static inline void ice_remove_arfs(struct ice_pf *pf) { }
diff --git a/drivers/net/ethernet/intel/ice/ice_base.c b/drivers/net/ethernet/intel/ice/ice_base.c
index b3234a55a253..6db4ad8fc70b 100644
--- a/drivers/net/ethernet/intel/ice/ice_base.c
+++ b/drivers/net/ethernet/intel/ice/ice_base.c
@@ -147,10 +147,6 @@ skip_alloc:
 	q_vector->reg_idx = q_vector->irq.index;
 	q_vector->vf_reg_idx = q_vector->irq.index;
 
-	/* only set affinity_mask if the CPU is online */
-	if (cpu_online(v_idx))
-		cpumask_set_cpu(v_idx, &q_vector->affinity_mask);
-
 	/* This will not be called in the driver load path because the netdev
 	 * will not be created yet. All other cases with register the NAPI
 	 * handler here (i.e. resume, reset/rebuild, etc.)
@@ -276,7 +272,8 @@ static void ice_cfg_xps_tx_ring(struct ice_tx_ring *ring)
 	if (test_and_set_bit(ICE_TX_XPS_INIT_DONE, ring->xps_state))
 		return;
 
-	netif_set_xps_queue(ring->netdev, &ring->q_vector->affinity_mask,
+	netif_set_xps_queue(ring->netdev,
+			    &ring->q_vector->napi.config->affinity_mask,
 			    ring->q_index);
 }
 
diff --git a/drivers/net/ethernet/intel/ice/ice_lib.c b/drivers/net/ethernet/intel/ice/ice_lib.c
index ce30674abf8f..715efd8a359f 100644
--- a/drivers/net/ethernet/intel/ice/ice_lib.c
+++ b/drivers/net/ethernet/intel/ice/ice_lib.c
@@ -2592,7 +2592,6 @@ void ice_vsi_free_irq(struct ice_vsi *vsi)
 		return;
 
 	vsi->irqs_ready = false;
-	ice_free_cpu_rx_rmap(vsi);
 
 	ice_for_each_q_vector(vsi, i) {
 		int irq_num;
@@ -2605,12 +2604,6 @@ void ice_vsi_free_irq(struct ice_vsi *vsi)
 		      vsi->q_vectors[i]->num_ring_rx))
 			continue;
 
-		/* clear the affinity notifier in the IRQ descriptor */
-		if (!IS_ENABLED(CONFIG_RFS_ACCEL))
-			irq_set_affinity_notifier(irq_num, NULL);
-
-		/* clear the affinity_hint in the IRQ descriptor */
-		irq_update_affinity_hint(irq_num, NULL);
 		synchronize_irq(irq_num);
 		devm_free_irq(ice_pf_to_dev(pf), irq_num, vsi->q_vectors[i]);
 	}
diff --git a/drivers/net/ethernet/intel/ice/ice_main.c b/drivers/net/ethernet/intel/ice/ice_main.c
index b084839eb811..eff4afabeef6 100644
--- a/drivers/net/ethernet/intel/ice/ice_main.c
+++ b/drivers/net/ethernet/intel/ice/ice_main.c
@@ -2527,34 +2527,6 @@ int ice_schedule_reset(struct ice_pf *pf, enum ice_reset_req reset)
 	return 0;
 }
 
-/**
- * ice_irq_affinity_notify - Callback for affinity changes
- * @notify: context as to what irq was changed
- * @mask: the new affinity mask
- *
- * This is a callback function used by the irq_set_affinity_notifier function
- * so that we may register to receive changes to the irq affinity masks.
- */
-static void
-ice_irq_affinity_notify(struct irq_affinity_notify *notify,
-			const cpumask_t *mask)
-{
-	struct ice_q_vector *q_vector =
-		container_of(notify, struct ice_q_vector, affinity_notify);
-
-	cpumask_copy(&q_vector->affinity_mask, mask);
-}
-
-/**
- * ice_irq_affinity_release - Callback for affinity notifier release
- * @ref: internal core kernel usage
- *
- * This is a callback function used by the irq_set_affinity_notifier function
- * to inform the current notification subscriber that they will no longer
- * receive notifications.
- */
-static void ice_irq_affinity_release(struct kref __always_unused *ref) {}
-
 /**
  * ice_vsi_ena_irq - Enable IRQ for the given VSI
  * @vsi: the VSI being configured
@@ -2618,19 +2590,6 @@ static int ice_vsi_req_irq_msix(struct ice_vsi *vsi, char *basename)
 				   err);
 			goto free_q_irqs;
 		}
-
-		/* register for affinity change notifications */
-		if (!IS_ENABLED(CONFIG_RFS_ACCEL)) {
-			struct irq_affinity_notify *affinity_notify;
-
-			affinity_notify = &q_vector->affinity_notify;
-			affinity_notify->notify = ice_irq_affinity_notify;
-			affinity_notify->release = ice_irq_affinity_release;
-			irq_set_affinity_notifier(irq_num, affinity_notify);
-		}
-
-		/* assign the mask for this irq */
-		irq_update_affinity_hint(irq_num, &q_vector->affinity_mask);
 	}
 
 	err = ice_set_cpu_rx_rmap(vsi);
@@ -2646,9 +2605,6 @@ static int ice_vsi_req_irq_msix(struct ice_vsi *vsi, char *basename)
 free_q_irqs:
 	while (vector--) {
 		irq_num = vsi->q_vectors[vector]->irq.virq;
-		if (!IS_ENABLED(CONFIG_RFS_ACCEL))
-			irq_set_affinity_notifier(irq_num, NULL);
-		irq_update_affinity_hint(irq_num, NULL);
 		devm_free_irq(dev, irq_num, &vsi->q_vectors[vector]);
 	}
 	return err;
@@ -3675,6 +3631,9 @@ void ice_set_netdev_features(struct net_device *netdev)
 	 */
 	netdev->hw_features |= NETIF_F_RXFCS;
 
+	/* Allow core to manage IRQs affinity */
+	netif_set_affinity_auto(netdev);
+
 	netif_set_tso_max_size(netdev, ICE_MAX_TSO_SIZE);
 }
 
-- 
2.51.0


From deab38f8f0118615feed2cd95da5f17aa9bff7ef Mon Sep 17 00:00:00 2001
From: Ahmed Zaki <ahmed.zaki@intel.com>
Date: Mon, 24 Feb 2025 16:22:26 -0700
Subject: [PATCH 14/16] idpf: use napi's irq affinity

Delete the driver CPU affinity info and use the core's napi config
instead.

Signed-off-by: Ahmed Zaki <ahmed.zaki@intel.com>
Link: https://patch.msgid.link/20250224232228.990783-6-ahmed.zaki@intel.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 drivers/net/ethernet/intel/idpf/idpf_lib.c  |  1 +
 drivers/net/ethernet/intel/idpf/idpf_txrx.c | 22 +++++++--------------
 drivers/net/ethernet/intel/idpf/idpf_txrx.h |  6 ++----
 3 files changed, 10 insertions(+), 19 deletions(-)

diff --git a/drivers/net/ethernet/intel/idpf/idpf_lib.c b/drivers/net/ethernet/intel/idpf/idpf_lib.c
index a3d6b8f198a8..f3aea7bcdaa3 100644
--- a/drivers/net/ethernet/intel/idpf/idpf_lib.c
+++ b/drivers/net/ethernet/intel/idpf/idpf_lib.c
@@ -814,6 +814,7 @@ static int idpf_cfg_netdev(struct idpf_vport *vport)
 	netdev->hw_features |= dflt_features | offloads;
 	netdev->hw_enc_features |= dflt_features | offloads;
 	idpf_set_ethtool_ops(netdev);
+	netif_set_affinity_auto(netdev);
 	SET_NETDEV_DEV(netdev, &adapter->pdev->dev);
 
 	/* carrier off on init to avoid Tx hangs */
diff --git a/drivers/net/ethernet/intel/idpf/idpf_txrx.c b/drivers/net/ethernet/intel/idpf/idpf_txrx.c
index 2747dc69999a..672dfad1fb21 100644
--- a/drivers/net/ethernet/intel/idpf/idpf_txrx.c
+++ b/drivers/net/ethernet/intel/idpf/idpf_txrx.c
@@ -3551,8 +3551,6 @@ void idpf_vport_intr_rel(struct idpf_vport *vport)
 		q_vector->tx = NULL;
 		kfree(q_vector->rx);
 		q_vector->rx = NULL;
-
-		free_cpumask_var(q_vector->affinity_mask);
 	}
 
 	kfree(vport->q_vectors);
@@ -3579,8 +3577,6 @@ static void idpf_vport_intr_rel_irq(struct idpf_vport *vport)
 		vidx = vport->q_vector_idxs[vector];
 		irq_num = adapter->msix_entries[vidx].vector;
 
-		/* clear the affinity_mask in the IRQ descriptor */
-		irq_set_affinity_hint(irq_num, NULL);
 		kfree(free_irq(irq_num, q_vector));
 	}
 }
@@ -3768,8 +3764,6 @@ static int idpf_vport_intr_req_irq(struct idpf_vport *vport)
 				   "Request_irq failed, error: %d\n", err);
 			goto free_q_irqs;
 		}
-		/* assign the mask for this irq */
-		irq_set_affinity_hint(irq_num, q_vector->affinity_mask);
 	}
 
 	return 0;
@@ -4181,7 +4175,8 @@ static int idpf_vport_intr_init_vec_idx(struct idpf_vport *vport)
 static void idpf_vport_intr_napi_add_all(struct idpf_vport *vport)
 {
 	int (*napi_poll)(struct napi_struct *napi, int budget);
-	u16 v_idx;
+	u16 v_idx, qv_idx;
+	int irq_num;
 
 	if (idpf_is_queue_model_split(vport->txq_model))
 		napi_poll = idpf_vport_splitq_napi_poll;
@@ -4190,12 +4185,12 @@ static void idpf_vport_intr_napi_add_all(struct idpf_vport *vport)
 
 	for (v_idx = 0; v_idx < vport->num_q_vectors; v_idx++) {
 		struct idpf_q_vector *q_vector = &vport->q_vectors[v_idx];
+		qv_idx = vport->q_vector_idxs[v_idx];
+		irq_num = vport->adapter->msix_entries[qv_idx].vector;
 
-		netif_napi_add(vport->netdev, &q_vector->napi, napi_poll);
-
-		/* only set affinity_mask if the CPU is online */
-		if (cpu_online(v_idx))
-			cpumask_set_cpu(v_idx, q_vector->affinity_mask);
+		netif_napi_add_config(vport->netdev, &q_vector->napi,
+				      napi_poll, v_idx);
+		netif_napi_set_irq(&q_vector->napi, irq_num);
 	}
 }
 
@@ -4239,9 +4234,6 @@ int idpf_vport_intr_alloc(struct idpf_vport *vport)
 		q_vector->rx_intr_mode = IDPF_ITR_DYNAMIC;
 		q_vector->rx_itr_idx = VIRTCHNL2_ITR_IDX_0;
 
-		if (!zalloc_cpumask_var(&q_vector->affinity_mask, GFP_KERNEL))
-			goto error;
-
 		q_vector->tx = kcalloc(txqs_per_vector, sizeof(*q_vector->tx),
 				       GFP_KERNEL);
 		if (!q_vector->tx)
diff --git a/drivers/net/ethernet/intel/idpf/idpf_txrx.h b/drivers/net/ethernet/intel/idpf/idpf_txrx.h
index cd9a20c9cc7f..b029f566e57c 100644
--- a/drivers/net/ethernet/intel/idpf/idpf_txrx.h
+++ b/drivers/net/ethernet/intel/idpf/idpf_txrx.h
@@ -382,7 +382,6 @@ struct idpf_intr_reg {
  * @rx_intr_mode: Dynamic ITR or not
  * @rx_itr_idx: RX ITR index
  * @v_idx: Vector index
- * @affinity_mask: CPU affinity mask
  */
 struct idpf_q_vector {
 	__cacheline_group_begin_aligned(read_mostly);
@@ -419,13 +418,12 @@ struct idpf_q_vector {
 	__cacheline_group_begin_aligned(cold);
 	u16 v_idx;
 
-	cpumask_var_t affinity_mask;
 	__cacheline_group_end_aligned(cold);
 };
 libeth_cacheline_set_assert(struct idpf_q_vector, 120,
 			    24 + sizeof(struct napi_struct) +
 			    2 * sizeof(struct dim),
-			    8 + sizeof(cpumask_var_t));
+			    8);
 
 struct idpf_rx_queue_stats {
 	u64_stats_t packets;
@@ -921,7 +919,7 @@ static inline int idpf_q_vector_to_mem(const struct idpf_q_vector *q_vector)
 	if (!q_vector)
 		return NUMA_NO_NODE;
 
-	cpu = cpumask_first(q_vector->affinity_mask);
+	cpu = cpumask_first(&q_vector->napi.config->affinity_mask);
 
 	return cpu < nr_cpu_ids ? cpu_to_mem(cpu) : NUMA_NO_NODE;
 }
-- 
2.51.0


From 185646a8a0a88c9016a782f9b98eea0eb6078512 Mon Sep 17 00:00:00 2001
From: Jakub Kicinski <kuba@kernel.org>
Date: Mon, 24 Feb 2025 16:22:27 -0700
Subject: [PATCH 15/16] selftests: drv-net: add tests for napi IRQ affinity
 notifiers

Add tests to check that the napi retained the IRQ after down/up,
multiple changes in the number of rx queues and after
attaching/releasing XDP program.

Tested on ice and idpf:

   # NETIF=<iface> tools/testing/selftests/drivers/net/hw/irq.py
    KTAP version 1
    1..4
    ok 1 irq.check_irqs_reported
    ok 2 irq.check_reconfig_queues
    ok 3 irq.check_reconfig_xdp
    ok 4 irq.check_down
    # Totals: pass:4 fail:0 xfail:0 xpass:0 skip:0 error:0

Tested-by: Ahmed Zaki <ahmed.zaki@intel.com>
Link: https://patch.msgid.link/20250224232228.990783-7-ahmed.zaki@intel.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
---
 .../testing/selftests/drivers/net/hw/Makefile |  4 +
 tools/testing/selftests/drivers/net/hw/irq.py | 99 +++++++++++++++++++
 .../selftests/drivers/net/hw/xdp_dummy.bpf.c  | 13 +++
 .../selftests/drivers/net/lib/py/env.py       |  8 +-
 4 files changed, 123 insertions(+), 1 deletion(-)
 create mode 100755 tools/testing/selftests/drivers/net/hw/irq.py
 create mode 100644 tools/testing/selftests/drivers/net/hw/xdp_dummy.bpf.c

diff --git a/tools/testing/selftests/drivers/net/hw/Makefile b/tools/testing/selftests/drivers/net/hw/Makefile
index b2be14d0cc56..cde5814ff9a7 100644
--- a/tools/testing/selftests/drivers/net/hw/Makefile
+++ b/tools/testing/selftests/drivers/net/hw/Makefile
@@ -10,6 +10,7 @@ TEST_PROGS = \
 	ethtool_rmon.sh \
 	hw_stats_l3.sh \
 	hw_stats_l3_gre.sh \
+	irq.py \
 	loopback.sh \
 	nic_link_layer.py \
 	nic_performance.py \
@@ -34,9 +35,12 @@ TEST_INCLUDES := \
 # YNL files, must be before "include ..lib.mk"
 YNL_GEN_FILES := ncdevmem
 TEST_GEN_FILES += $(YNL_GEN_FILES)
+TEST_GEN_FILES += $(patsubst %.c,%.o,$(wildcard *.bpf.c))
 
 include ../../../lib.mk
 
 # YNL build
 YNL_GENS := ethtool netdev
 include ../../../net/ynl.mk
+
+include ../../../net/bpf.mk
diff --git a/tools/testing/selftests/drivers/net/hw/irq.py b/tools/testing/selftests/drivers/net/hw/irq.py
new file mode 100755
index 000000000000..42ab98370245
--- /dev/null
+++ b/tools/testing/selftests/drivers/net/hw/irq.py
@@ -0,0 +1,99 @@
+#!/usr/bin/env python3
+# SPDX-License-Identifier: GPL-2.0
+
+from lib.py import ksft_run, ksft_exit
+from lib.py import ksft_ge, ksft_eq
+from lib.py import KsftSkipEx
+from lib.py import ksft_disruptive
+from lib.py import EthtoolFamily, NetdevFamily
+from lib.py import NetDrvEnv
+from lib.py import cmd, ip, defer
+
+
+def read_affinity(irq) -> str:
+    with open(f'/proc/irq/{irq}/smp_affinity', 'r') as fp:
+        return fp.read().lstrip("0,").strip()
+
+
+def write_affinity(irq, what) -> str:
+    if what != read_affinity(irq):
+        with open(f'/proc/irq/{irq}/smp_affinity', 'w') as fp:
+            fp.write(what)
+
+
+def check_irqs_reported(cfg) -> None:
+    """ Check that device reports IRQs for NAPI instances """
+    napis = cfg.netnl.napi_get({"ifindex": cfg.ifindex}, dump=True)
+    irqs = sum(['irq' in x for x in napis])
+
+    ksft_ge(irqs, 1)
+    ksft_eq(irqs, len(napis))
+
+
+def _check_reconfig(cfg, reconfig_cb) -> None:
+    napis = cfg.netnl.napi_get({"ifindex": cfg.ifindex}, dump=True)
+    for n in reversed(napis):
+        if 'irq' in n:
+            break
+    else:
+        raise KsftSkipEx(f"Device has no NAPI with IRQ attribute (#napis: {len(napis)}")
+
+    old = read_affinity(n['irq'])
+    # pick an affinity that's not the current one
+    new = "3" if old != "3" else "5"
+    write_affinity(n['irq'], new)
+    defer(write_affinity, n['irq'], old)
+
+    reconfig_cb(cfg)
+
+    ksft_eq(read_affinity(n['irq']), new, comment="IRQ affinity changed after reconfig")
+
+
+def check_reconfig_queues(cfg) -> None:
+    def reconfig(cfg) -> None:
+        channels = cfg.ethnl.channels_get({'header': {'dev-index': cfg.ifindex}})
+        if channels['combined-count'] == 0:
+            rx_type = 'rx'
+        else:
+            rx_type = 'combined'
+        cur_queue_cnt = channels[f'{rx_type}-count']
+        max_queue_cnt = channels[f'{rx_type}-max']
+
+        cmd(f"ethtool -L {cfg.ifname} {rx_type} 1")
+        cmd(f"ethtool -L {cfg.ifname} {rx_type} {max_queue_cnt}")
+        cmd(f"ethtool -L {cfg.ifname} {rx_type} {cur_queue_cnt}")
+
+    _check_reconfig(cfg, reconfig)
+
+
+def check_reconfig_xdp(cfg) -> None:
+    def reconfig(cfg) -> None:
+        ip(f"link set dev %s xdp obj %s sec xdp" %
+            (cfg.ifname, cfg.rpath("xdp_dummy.bpf.o")))
+        ip(f"link set dev %s xdp off" % cfg.ifname)
+
+    _check_reconfig(cfg, reconfig)
+
+
+@ksft_disruptive
+def check_down(cfg) -> None:
+    def reconfig(cfg) -> None:
+        ip("link set dev %s down" % cfg.ifname)
+        ip("link set dev %s up" % cfg.ifname)
+
+    _check_reconfig(cfg, reconfig)
+
+
+def main() -> None:
+    with NetDrvEnv(__file__, nsim_test=False) as cfg:
+        cfg.ethnl = EthtoolFamily()
+        cfg.netnl = NetdevFamily()
+
+        ksft_run([check_irqs_reported, check_reconfig_queues,
+                  check_reconfig_xdp, check_down],
+                 args=(cfg, ))
+    ksft_exit()
+
+
+if __name__ == "__main__":
+    main()
diff --git a/tools/testing/selftests/drivers/net/hw/xdp_dummy.bpf.c b/tools/testing/selftests/drivers/net/hw/xdp_dummy.bpf.c
new file mode 100644
index 000000000000..d988b2e0cee8
--- /dev/null
+++ b/tools/testing/selftests/drivers/net/hw/xdp_dummy.bpf.c
@@ -0,0 +1,13 @@
+// SPDX-License-Identifier: GPL-2.0
+
+#define KBUILD_MODNAME "xdp_dummy"
+#include <linux/bpf.h>
+#include <bpf/bpf_helpers.h>
+
+SEC("xdp")
+int xdp_dummy_prog(struct xdp_md *ctx)
+{
+	return XDP_PASS;
+}
+
+char _license[] SEC("license") = "GPL";
diff --git a/tools/testing/selftests/drivers/net/lib/py/env.py b/tools/testing/selftests/drivers/net/lib/py/env.py
index 96b33b5ef9dd..fd4d674e6c72 100644
--- a/tools/testing/selftests/drivers/net/lib/py/env.py
+++ b/tools/testing/selftests/drivers/net/lib/py/env.py
@@ -58,14 +58,20 @@ class NetDrvEnv(NetDrvEnvBase):
     """
     Class for a single NIC / host env, with no remote end
     """
-    def __init__(self, src_path, **kwargs):
+    def __init__(self, src_path, nsim_test=None, **kwargs):
         super().__init__(src_path)
 
         self._ns = None
 
         if 'NETIF' in self.env:
+            if nsim_test is True:
+                raise KsftXfailEx("Test only works on netdevsim")
+
             self.dev = ip("-d link show dev " + self.env['NETIF'], json=True)[0]
         else:
+            if nsim_test is False:
+                raise KsftXfailEx("Test does not work on netdevsim")
+
             self._ns = NetdevSimDev(**kwargs)
             self.dev = self._ns.nsims[0].dev
         self.ifname = self.dev['ifname']
-- 
2.51.0


From af4a5da8ed54561b7e856534cf62ea07632e7b7b Mon Sep 17 00:00:00 2001
From: Arnd Bergmann <arnd@arndb.de>
Date: Tue, 25 Feb 2025 09:57:14 +0100
Subject: [PATCH 16/16] pktgen: avoid unused-const-variable warning

When extra warnings are enable, there are configurations that build
pktgen without CONFIG_XFRM, which leaves a static const variable unused:

net/core/pktgen.c:213:1: error: unused variable 'F_IPSEC' [-Werror,-Wunused-const-variable]
  213 | PKT_FLAGS
      | ^~~~~~~~~
net/core/pktgen.c:197:2: note: expanded from macro 'PKT_FLAGS'
  197 |         pf(IPSEC)               /* ipsec on for flows */                \
      |         ^~~~~~~~~

This could be marked as __maybe_unused, or by making the one use visible
to the compiler by slightly rearranging the #ifdef blocks. The second
variant looks slightly nicer here, so use that.

Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Reviewed-by: Peter Seiderer <ps.report@gmx.net>
Link: https://patch.msgid.link/20250225085722.469868-1-arnd@kernel.org
Signed-off-by: Paolo Abeni <pabeni@redhat.com>
---
 net/core/pktgen.c | 9 ++-------
 1 file changed, 2 insertions(+), 7 deletions(-)

diff --git a/net/core/pktgen.c b/net/core/pktgen.c
index 55064713223e..402e01a2ce19 100644
--- a/net/core/pktgen.c
+++ b/net/core/pktgen.c
@@ -158,9 +158,7 @@
 #include <net/udp.h>
 #include <net/ip6_checksum.h>
 #include <net/addrconf.h>
-#ifdef CONFIG_XFRM
 #include <net/xfrm.h>
-#endif
 #include <net/netns/generic.h>
 #include <asm/byteorder.h>
 #include <linux/rcupdate.h>
@@ -2363,13 +2361,13 @@ static inline int f_pick(struct pktgen_dev *pkt_dev)
 }
 
 
-#ifdef CONFIG_XFRM
 /* If there was already an IPSEC SA, we keep it as is, else
  * we go look for it ...
 */
 #define DUMMY_MARK 0
 static void get_ipsec_sa(struct pktgen_dev *pkt_dev, int flow)
 {
+#ifdef CONFIG_XFRM
 	struct xfrm_state *x = pkt_dev->flows[flow].x;
 	struct pktgen_net *pn = net_generic(dev_net(pkt_dev->odev), pg_net_id);
 	if (!x) {
@@ -2395,11 +2393,10 @@ static void get_ipsec_sa(struct pktgen_dev *pkt_dev, int flow)
 		}
 
 	}
-}
 #endif
+}
 static void set_cur_queue_map(struct pktgen_dev *pkt_dev)
 {
-
 	if (pkt_dev->flags & F_QUEUE_MAP_CPU)
 		pkt_dev->cur_queue_map = smp_processor_id();
 
@@ -2574,10 +2571,8 @@ static void mod_cur_headers(struct pktgen_dev *pkt_dev)
 				pkt_dev->flows[flow].flags |= F_INIT;
 				pkt_dev->flows[flow].cur_daddr =
 				    pkt_dev->cur_daddr;
-#ifdef CONFIG_XFRM
 				if (pkt_dev->flags & F_IPSEC)
 					get_ipsec_sa(pkt_dev, flow);
-#endif
 				pkt_dev->nflows++;
 			}
 		}
-- 
2.51.0