stable.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To: stable@vger.kernel.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	patches@lists.linux.dev, Eric Dumazet <edumazet@google.com>,
	Simon Horman <simon.horman@corigine.com>,
	Kuniyuki Iwashima <kuniyu@amazon.com>,
	"David S. Miller" <davem@davemloft.net>,
	Sasha Levin <sashal@kernel.org>
Subject: [PATCH 5.10 26/68] rfs: annotate lockless accesses to sk->sk_rxhash
Date: Mon, 12 Jun 2023 12:26:18 +0200	[thread overview]
Message-ID: <20230612101659.507635367@linuxfoundation.org> (raw)
In-Reply-To: <20230612101658.437327280@linuxfoundation.org>

From: Eric Dumazet <edumazet@google.com>

[ Upstream commit 1e5c647c3f6d4f8497dedcd226204e1880e0ffb3 ]

Add READ_ONCE()/WRITE_ONCE() on accesses to sk->sk_rxhash.

This also prevents a (smart ?) compiler to remove the condition in:

if (sk->sk_rxhash != newval)
	sk->sk_rxhash = newval;

We need the condition to avoid dirtying a shared cache line.

Fixes: fec5e652e58f ("rfs: Receive Flow Steering")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reviewed-by: Simon Horman <simon.horman@corigine.com>
Reviewed-by: Kuniyuki Iwashima <kuniyu@amazon.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 include/net/sock.h | 18 +++++++++++++-----
 1 file changed, 13 insertions(+), 5 deletions(-)

diff --git a/include/net/sock.h b/include/net/sock.h
index 3da0601b573ed..51b499d745499 100644
--- a/include/net/sock.h
+++ b/include/net/sock.h
@@ -1073,8 +1073,12 @@ static inline void sock_rps_record_flow(const struct sock *sk)
 		 * OR	an additional socket flag
 		 * [1] : sk_state and sk_prot are in the same cache line.
 		 */
-		if (sk->sk_state == TCP_ESTABLISHED)
-			sock_rps_record_flow_hash(sk->sk_rxhash);
+		if (sk->sk_state == TCP_ESTABLISHED) {
+			/* This READ_ONCE() is paired with the WRITE_ONCE()
+			 * from sock_rps_save_rxhash() and sock_rps_reset_rxhash().
+			 */
+			sock_rps_record_flow_hash(READ_ONCE(sk->sk_rxhash));
+		}
 	}
 #endif
 }
@@ -1083,15 +1087,19 @@ static inline void sock_rps_save_rxhash(struct sock *sk,
 					const struct sk_buff *skb)
 {
 #ifdef CONFIG_RPS
-	if (unlikely(sk->sk_rxhash != skb->hash))
-		sk->sk_rxhash = skb->hash;
+	/* The following WRITE_ONCE() is paired with the READ_ONCE()
+	 * here, and another one in sock_rps_record_flow().
+	 */
+	if (unlikely(READ_ONCE(sk->sk_rxhash) != skb->hash))
+		WRITE_ONCE(sk->sk_rxhash, skb->hash);
 #endif
 }
 
 static inline void sock_rps_reset_rxhash(struct sock *sk)
 {
 #ifdef CONFIG_RPS
-	sk->sk_rxhash = 0;
+	/* Paired with READ_ONCE() in sock_rps_record_flow() */
+	WRITE_ONCE(sk->sk_rxhash, 0);
 #endif
 }
 
-- 
2.39.2




  parent reply	other threads:[~2023-06-12 10:50 UTC|newest]

Thread overview: 79+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-06-12 10:25 [PATCH 5.10 00/68] 5.10.184-rc1 review Greg Kroah-Hartman
2023-06-12 10:25 ` [PATCH 5.10 01/68] ata: ahci: fix enum constants for gcc-13 Greg Kroah-Hartman
2023-06-12 10:25 ` [PATCH 5.10 02/68] gcc-plugins: Reorganize gimple includes for GCC 13 Greg Kroah-Hartman
2023-06-12 10:25 ` [PATCH 5.10 03/68] sfc (gcc13): synchronize ef100_enqueue_skb()s return type Greg Kroah-Hartman
2023-06-12 10:25 ` [PATCH 5.10 04/68] remove the sx8 block driver Greg Kroah-Hartman
2023-06-12 10:25 ` [PATCH 5.10 05/68] bonding (gcc13): synchronize bond_{a,t}lb_xmit() types Greg Kroah-Hartman
2023-06-12 10:25 ` [PATCH 5.10 06/68] f2fs: fix iostat lock protection Greg Kroah-Hartman
2023-06-12 10:25 ` [PATCH 5.10 07/68] blk-iocost: avoid 64-bit division in ioc_timer_fn Greg Kroah-Hartman
2023-06-12 10:26 ` [PATCH 5.10 08/68] block/blk-iocost (gcc13): keep large values in a new enum Greg Kroah-Hartman
2023-06-12 10:26 ` [PATCH 5.10 09/68] i40iw: fix build warning in i40iw_manage_apbvt() Greg Kroah-Hartman
2023-06-12 10:26 ` [PATCH 5.10 10/68] i40e: fix build warnings in i40e_alloc.h Greg Kroah-Hartman
2023-06-12 10:26 ` [PATCH 5.10 11/68] i40e: fix build warning in ice_fltr_add_mac_to_list() Greg Kroah-Hartman
2023-06-12 10:26 ` [PATCH 5.10 12/68] staging: vchiq_core: drop vchiq_status from vchiq_initialise Greg Kroah-Hartman
2023-06-12 10:26 ` [PATCH 5.10 13/68] spi: qup: Request DMA before enabling clocks Greg Kroah-Hartman
2023-06-12 10:26 ` [PATCH 5.10 14/68] afs: Fix setting of mtime when creating a file/dir/symlink Greg Kroah-Hartman
2023-06-12 10:26 ` [PATCH 5.10 15/68] wifi: mt76: mt7615: fix possible race in mt7615_mac_sta_poll Greg Kroah-Hartman
2023-06-12 10:26 ` [PATCH 5.10 16/68] neighbour: fix unaligned access to pneigh_entry Greg Kroah-Hartman
2023-06-12 10:26 ` [PATCH 5.10 17/68] net: dsa: lan9303: allow vid != 0 in port_fdb_{add|del} methods Greg Kroah-Hartman
2023-06-12 10:26 ` [PATCH 5.10 18/68] net/smc: Avoid to access invalid RMBs MRs in SMCRv1 ADD LINK CONT Greg Kroah-Hartman
2023-06-12 10:26 ` [PATCH 5.10 19/68] net/sched: fq_pie: ensure reasonable TCA_FQ_PIE_QUANTUM values Greg Kroah-Hartman
2023-06-12 10:26 ` [PATCH 5.10 20/68] Bluetooth: Fix l2cap_disconnect_req deadlock Greg Kroah-Hartman
2023-06-12 10:26 ` [PATCH 5.10 21/68] Bluetooth: L2CAP: Add missing checks for invalid DCID Greg Kroah-Hartman
2023-06-12 10:26 ` [PATCH 5.10 22/68] qed/qede: Fix scheduling while atomic Greg Kroah-Hartman
2023-06-12 10:26 ` [PATCH 5.10 23/68] netfilter: conntrack: fix NULL pointer dereference in nf_confirm_cthelper Greg Kroah-Hartman
2023-06-12 10:26 ` [PATCH 5.10 24/68] netfilter: ipset: Add schedule point in call_ad() Greg Kroah-Hartman
2023-06-12 10:26 ` [PATCH 5.10 25/68] ipv6: rpl: Fix Route of Death Greg Kroah-Hartman
2023-06-12 10:26 ` Greg Kroah-Hartman [this message]
2023-06-12 10:26 ` [PATCH 5.10 27/68] rfs: annotate lockless accesses to RFS sock flow table Greg Kroah-Hartman
2023-06-12 10:26 ` [PATCH 5.10 28/68] net: sched: move rtm_tca_policy declaration to include file Greg Kroah-Hartman
2023-06-12 10:26 ` [PATCH 5.10 29/68] net: sched: fix possible refcount leak in tc_chain_tmplt_add() Greg Kroah-Hartman
2023-06-12 10:26 ` [PATCH 5.10 30/68] bpf: Add extra path pointer check to d_path helper Greg Kroah-Hartman
2023-06-12 10:26 ` [PATCH 5.10 31/68] lib: cpu_rmap: Fix potential use-after-free in irq_cpu_rmap_release() Greg Kroah-Hartman
2023-06-12 10:26 ` [PATCH 5.10 32/68] bnxt_en: Dont issue AP reset during ethtools reset operation Greg Kroah-Hartman
2023-06-12 10:26 ` [PATCH 5.10 33/68] bnxt_en: Query default VLAN before VNIC setup on a VF Greg Kroah-Hartman
2023-06-12 10:26 ` [PATCH 5.10 34/68] bnxt_en: Implement .set_port / .unset_port UDP tunnel callbacks Greg Kroah-Hartman
2023-06-12 10:26 ` [PATCH 5.10 35/68] batman-adv: Broken sync while rescheduling delayed work Greg Kroah-Hartman
2023-06-12 10:26 ` [PATCH 5.10 36/68] Input: xpad - delete a Razer DeathAdder mouse VID/PID entry Greg Kroah-Hartman
2023-06-12 10:26 ` [PATCH 5.10 37/68] Input: psmouse - fix OOB access in Elantech protocol Greg Kroah-Hartman
2023-06-12 10:26 ` [PATCH 5.10 38/68] ALSA: hda/realtek: Add a quirk for HP Slim Desktop S01 Greg Kroah-Hartman
2023-06-12 10:26 ` [PATCH 5.10 39/68] ALSA: hda/realtek: Add Lenovo P3 Tower platform Greg Kroah-Hartman
2023-06-12 10:26 ` [PATCH 5.10 40/68] drm/amdgpu: fix xclk freq on CHIP_STONEY Greg Kroah-Hartman
2023-06-12 10:26 ` [PATCH 5.10 41/68] can: j1939: j1939_sk_send_loop_abort(): improved error queue handling in J1939 Socket Greg Kroah-Hartman
2023-06-12 10:26 ` [PATCH 5.10 42/68] can: j1939: change j1939_netdev_lock type to mutex Greg Kroah-Hartman
2023-06-12 10:26 ` [PATCH 5.10 43/68] can: j1939: avoid possible use-after-free when j1939_can_rx_register fails Greg Kroah-Hartman
2023-06-12 10:26 ` [PATCH 5.10 44/68] ceph: fix use-after-free bug for inodes when flushing capsnaps Greg Kroah-Hartman
2023-06-12 10:26 ` [PATCH 5.10 45/68] s390/dasd: Use correct lock while counting channel queue length Greg Kroah-Hartman
2023-06-12 10:26 ` [PATCH 5.10 46/68] Bluetooth: Fix use-after-free in hci_remove_ltk/hci_remove_irk Greg Kroah-Hartman
2023-06-12 10:26 ` [PATCH 5.10 47/68] Bluetooth: hci_qca: fix debugfs registration Greg Kroah-Hartman
2023-06-12 10:26 ` [PATCH 5.10 48/68] tee: amdtee: Add return_origin to struct tee_cmd_load_ta Greg Kroah-Hartman
2023-06-12 10:26 ` [PATCH 5.10 49/68] rbd: move RBD_OBJ_FLAG_COPYUP_ENABLED flag setting Greg Kroah-Hartman
2023-06-12 10:26 ` [PATCH 5.10 50/68] rbd: get snapshot context after exclusive lock is ensured to be held Greg Kroah-Hartman
2023-06-12 10:26 ` [PATCH 5.10 51/68] pinctrl: meson-axg: add missing GPIOA_18 gpio group Greg Kroah-Hartman
2023-06-12 10:26 ` [PATCH 5.10 52/68] usb: usbfs: Enforce page requirements for mmap Greg Kroah-Hartman
2023-06-12 10:26 ` [PATCH 5.10 53/68] usb: usbfs: Use consistent mmap functions Greg Kroah-Hartman
2023-06-12 10:26 ` [PATCH 5.10 54/68] staging: vc04_services: fix gcc-13 build warning Greg Kroah-Hartman
2023-06-12 10:26 ` [PATCH 5.10 55/68] ASoC: codecs: wsa881x: do not set can_multi_write flag Greg Kroah-Hartman
2023-06-12 10:26 ` [PATCH 5.10 56/68] i2c: sprd: Delete i2c adapter in .removes error path Greg Kroah-Hartman
2023-06-12 10:26 ` [PATCH 5.10 57/68] eeprom: at24: also select REGMAP Greg Kroah-Hartman
2023-06-12 10:26 ` [PATCH 5.10 58/68] riscv: fix kprobe __user string arg print fault issue Greg Kroah-Hartman
2023-06-12 10:26 ` [PATCH 5.10 59/68] vhost: support PACKED when setting-getting vring_base Greg Kroah-Hartman
2023-06-12 10:26 ` [PATCH 5.10 60/68] Revert "ext4: dont clear SB_RDONLY when remounting r/w until quota is re-enabled" Greg Kroah-Hartman
2023-06-12 10:26 ` [PATCH 5.10 61/68] ext4: only check dquot_initialize_needed() when debugging Greg Kroah-Hartman
2023-06-12 10:26 ` [PATCH 5.10 62/68] tcp: fix tcp_min_tso_segs sysctl Greg Kroah-Hartman
2023-06-12 10:26 ` [PATCH 5.10 63/68] xfs: verify buffer contents when we skip log replay Greg Kroah-Hartman
2023-06-12 10:26 ` [PATCH 5.10 64/68] MIPS: locking/atomic: Fix atomic{_64,}_sub_if_positive Greg Kroah-Hartman
2023-06-12 10:26 ` [PATCH 5.10 65/68] drm/atomic: Dont pollute crtc_state->mode_blob with error pointers Greg Kroah-Hartman
2023-06-12 10:26 ` [PATCH 5.10 66/68] btrfs: check return value of btrfs_commit_transaction in relocation Greg Kroah-Hartman
2023-06-12 10:26 ` [PATCH 5.10 67/68] btrfs: unset reloc control if transaction commit fails in prepare_to_relocate() Greg Kroah-Hartman
2023-06-12 10:27 ` [PATCH 5.10 68/68] Revert "staging: rtl8192e: Replace macro RTL_PCI_DEVICE with PCI_DEVICE" Greg Kroah-Hartman
2023-06-12 21:52 ` [PATCH 5.10 00/68] 5.10.184-rc1 review Chris Paterson
2023-06-12 22:17 ` Shuah Khan
2023-06-13  8:38 ` Jon Hunter
2023-06-13  9:02 ` Naresh Kamboju
2023-06-13 12:13 ` Sudip Mukherjee (Codethink)
2023-06-13 16:35 ` Guenter Roeck
2023-06-13 17:36   ` Greg Kroah-Hartman
2023-06-13 18:01     ` Guenter Roeck
2023-06-13 17:45 ` Allen Pais
2023-06-13 23:09 ` Guenter Roeck

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=20230612101659.507635367@linuxfoundation.org \
    --to=gregkh@linuxfoundation.org \
    --cc=davem@davemloft.net \
    --cc=edumazet@google.com \
    --cc=kuniyu@amazon.com \
    --cc=patches@lists.linux.dev \
    --cc=sashal@kernel.org \
    --cc=simon.horman@corigine.com \
    --cc=stable@vger.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;
as well as URLs for NNTP newsgroup(s).