netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Eric Dumazet <edumazet@google.com>
To: "David S . Miller" <davem@davemloft.net>,
	Willem de Bruijn <willemb@google.com>,
	Mahesh Bandewar <maheshb@google.com>
Cc: netdev <netdev@vger.kernel.org>,
	Eric Dumazet <edumazet@google.com>,
	Eric Dumazet <eric.dumazet@gmail.com>
Subject: [PATCH net-next 7/8] net/packet: remove locking from packet_rcv_has_room()
Date: Wed, 12 Jun 2019 09:52:32 -0700	[thread overview]
Message-ID: <20190612165233.109749-8-edumazet@google.com> (raw)
In-Reply-To: <20190612165233.109749-1-edumazet@google.com>

__packet_rcv_has_room() can now be run without lock being held.

po->pressure is only a non persistent hint, we can mark
all read/write accesses with READ_ONCE()/WRITE_ONCE()
to document the fact that the field could be written
without any synchronization.

Signed-off-by: Eric Dumazet <edumazet@google.com>
---
 net/packet/af_packet.c | 20 +++++++++-----------
 1 file changed, 9 insertions(+), 11 deletions(-)

diff --git a/net/packet/af_packet.c b/net/packet/af_packet.c
index 860ca3e6abf5198214612e9acc095530b61dac40..d409e2fdaa7ee8ddf261354f91b682e403f40e9e 100644
--- a/net/packet/af_packet.c
+++ b/net/packet/af_packet.c
@@ -1260,15 +1260,13 @@ static int __packet_rcv_has_room(const struct packet_sock *po,
 
 static int packet_rcv_has_room(struct packet_sock *po, struct sk_buff *skb)
 {
-	int ret;
-	bool has_room;
+	int pressure, ret;
 
-	spin_lock_bh(&po->sk.sk_receive_queue.lock);
 	ret = __packet_rcv_has_room(po, skb);
-	has_room = ret == ROOM_NORMAL;
-	if (po->pressure == has_room)
-		po->pressure = !has_room;
-	spin_unlock_bh(&po->sk.sk_receive_queue.lock);
+	pressure = ret != ROOM_NORMAL;
+
+	if (READ_ONCE(po->pressure) != pressure)
+		WRITE_ONCE(po->pressure, pressure);
 
 	return ret;
 }
@@ -1353,7 +1351,7 @@ static unsigned int fanout_demux_rollover(struct packet_fanout *f,
 	i = j = min_t(int, po->rollover->sock, num - 1);
 	do {
 		po_next = pkt_sk(f->arr[i]);
-		if (po_next != po_skip && !po_next->pressure &&
+		if (po_next != po_skip && !READ_ONCE(po_next->pressure) &&
 		    packet_rcv_has_room(po_next, skb) == ROOM_NORMAL) {
 			if (i != j)
 				po->rollover->sock = i;
@@ -3310,7 +3308,7 @@ static int packet_recvmsg(struct socket *sock, struct msghdr *msg, size_t len,
 	if (skb == NULL)
 		goto out;
 
-	if (pkt_sk(sk)->pressure)
+	if (READ_ONCE(pkt_sk(sk)->pressure))
 		packet_rcv_has_room(pkt_sk(sk), NULL);
 
 	if (pkt_sk(sk)->has_vnet_hdr) {
@@ -4129,8 +4127,8 @@ static __poll_t packet_poll(struct file *file, struct socket *sock,
 			TP_STATUS_KERNEL))
 			mask |= EPOLLIN | EPOLLRDNORM;
 	}
-	if (po->pressure && __packet_rcv_has_room(po, NULL) == ROOM_NORMAL)
-		po->pressure = 0;
+	if (READ_ONCE(po->pressure) && __packet_rcv_has_room(po, NULL) == ROOM_NORMAL)
+		WRITE_ONCE(po->pressure, 0);
 	spin_unlock_bh(&sk->sk_receive_queue.lock);
 	spin_lock_bh(&sk->sk_write_queue.lock);
 	if (po->tx_ring.pg_vec) {
-- 
2.22.0.rc2.383.gf4fbbf30c2-goog


  parent reply	other threads:[~2019-06-12 16:53 UTC|newest]

Thread overview: 12+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-06-12 16:52 [PATCH net-next 0/8] net/packet: better behavior under DDOS Eric Dumazet
2019-06-12 16:52 ` [PATCH net-next 1/8] net/packet: constify __packet_get_status() argument Eric Dumazet
2019-06-12 16:52 ` [PATCH net-next 2/8] net/packet: constify packet_lookup_frame() and __tpacket_has_room() Eric Dumazet
2019-06-12 16:52 ` [PATCH net-next 3/8] net/packet: constify prb_lookup_block() and __tpacket_v3_has_room() Eric Dumazet
2019-06-12 16:52 ` [PATCH net-next 4/8] net/packet: constify __packet_rcv_has_room() Eric Dumazet
2019-06-12 16:52 ` [PATCH net-next 5/8] net/packet: make tp_drops atomic Eric Dumazet
2019-06-12 16:52 ` [PATCH net-next 6/8] net/packet: implement shortcut in tpacket_rcv() Eric Dumazet
2019-06-12 16:52 ` Eric Dumazet [this message]
2019-06-12 16:52 ` [PATCH net-next 8/8] net/packet: introduce packet_rcv_try_clear_pressure() helper Eric Dumazet
2019-06-13  0:11   ` Vinicius Costa Gomes
2019-06-12 17:15 ` [PATCH net-next 0/8] net/packet: better behavior under DDOS Willem de Bruijn
2019-06-15  1:53 ` David Miller

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=20190612165233.109749-8-edumazet@google.com \
    --to=edumazet@google.com \
    --cc=davem@davemloft.net \
    --cc=eric.dumazet@gmail.com \
    --cc=maheshb@google.com \
    --cc=netdev@vger.kernel.org \
    --cc=willemb@google.com \
    /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).