stable.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To: linux-kernel@vger.kernel.org
Cc: "Greg Kroah-Hartman" <gregkh@linuxfoundation.org>,
	stable@vger.kernel.org, "Eric Dumazet" <edumazet@google.com>,
	배석진 <soukjin.bae@samsung.com>,
	"David S. Miller" <davem@davemloft.net>
Subject: [PATCH 4.19 26/42] inet: frags: better deal with smp races
Date: Wed, 21 Nov 2018 20:06:02 +0100	[thread overview]
Message-ID: <20181121183149.133109771@linuxfoundation.org> (raw)
In-Reply-To: <20181121183147.869199006@linuxfoundation.org>

4.19-stable review patch.  If anyone has any objections, please let me know.

------------------

From: Eric Dumazet <edumazet@google.com>

[ Upstream commit 0d5b9311baf27bb545f187f12ecfd558220c607d ]

Multiple cpus might attempt to insert a new fragment in rhashtable,
if for example RPS is buggy, as reported by 배석진 in
https://patchwork.ozlabs.org/patch/994601/

We use rhashtable_lookup_get_insert_key() instead of
rhashtable_insert_fast() to let cpus losing the race
free their own inet_frag_queue and use the one that
was inserted by another cpu.

Fixes: 648700f76b03 ("inet: frags: use rhashtables for reassembly units")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reported-by: 배석진 <soukjin.bae@samsung.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
 net/ipv4/inet_fragment.c |   29 +++++++++++++++--------------
 1 file changed, 15 insertions(+), 14 deletions(-)

--- a/net/ipv4/inet_fragment.c
+++ b/net/ipv4/inet_fragment.c
@@ -178,21 +178,22 @@ static struct inet_frag_queue *inet_frag
 }
 
 static struct inet_frag_queue *inet_frag_create(struct netns_frags *nf,
-						void *arg)
+						void *arg,
+						struct inet_frag_queue **prev)
 {
 	struct inet_frags *f = nf->f;
 	struct inet_frag_queue *q;
-	int err;
 
 	q = inet_frag_alloc(nf, f, arg);
-	if (!q)
+	if (!q) {
+		*prev = ERR_PTR(-ENOMEM);
 		return NULL;
-
+	}
 	mod_timer(&q->timer, jiffies + nf->timeout);
 
-	err = rhashtable_insert_fast(&nf->rhashtable, &q->node,
-				     f->rhash_params);
-	if (err < 0) {
+	*prev = rhashtable_lookup_get_insert_key(&nf->rhashtable, &q->key,
+						 &q->node, f->rhash_params);
+	if (*prev) {
 		q->flags |= INET_FRAG_COMPLETE;
 		inet_frag_kill(q);
 		inet_frag_destroy(q);
@@ -204,22 +205,22 @@ static struct inet_frag_queue *inet_frag
 /* TODO : call from rcu_read_lock() and no longer use refcount_inc_not_zero() */
 struct inet_frag_queue *inet_frag_find(struct netns_frags *nf, void *key)
 {
-	struct inet_frag_queue *fq;
+	struct inet_frag_queue *fq = NULL, *prev;
 
 	if (!nf->high_thresh || frag_mem_limit(nf) > nf->high_thresh)
 		return NULL;
 
 	rcu_read_lock();
 
-	fq = rhashtable_lookup(&nf->rhashtable, key, nf->f->rhash_params);
-	if (fq) {
+	prev = rhashtable_lookup(&nf->rhashtable, key, nf->f->rhash_params);
+	if (!prev)
+		fq = inet_frag_create(nf, key, &prev);
+	if (prev && !IS_ERR(prev)) {
+		fq = prev;
 		if (!refcount_inc_not_zero(&fq->refcnt))
 			fq = NULL;
-		rcu_read_unlock();
-		return fq;
 	}
 	rcu_read_unlock();
-
-	return inet_frag_create(nf, key);
+	return fq;
 }
 EXPORT_SYMBOL(inet_frag_find);

  parent reply	other threads:[~2018-11-21 19:06 UTC|newest]

Thread overview: 58+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-11-21 19:05 [PATCH 4.19 00/42] 4.19.4-stable review Greg Kroah-Hartman
2018-11-21 19:05 ` [PATCH 4.19 01/42] flow_dissector: do not dissect l4 ports for fragments Greg Kroah-Hartman
2018-11-21 19:05 ` [PATCH 4.19 02/42] ibmvnic: fix accelerated VLAN handling Greg Kroah-Hartman
2018-11-21 19:05 ` [PATCH 4.19 03/42] ip_tunnel: dont force DF when MTU is locked Greg Kroah-Hartman
2018-11-21 19:05 ` [PATCH 4.19 04/42] ipv6: fix a dst leak when removing its exception Greg Kroah-Hartman
2018-11-21 19:05 ` [PATCH 4.19 05/42] ipv6: Fix PMTU updates for UDP/raw sockets in presence of VRF Greg Kroah-Hartman
2018-11-21 19:05 ` [PATCH 4.19 06/42] net: bcmgenet: protect stop from timeout Greg Kroah-Hartman
2018-11-21 19:05 ` [PATCH 4.19 07/42] net-gro: reset skb->pkt_type in napi_reuse_skb() Greg Kroah-Hartman
2018-11-21 19:05 ` [PATCH 4.19 08/42] sctp: not allow to set asoc prsctp_enable by sockopt Greg Kroah-Hartman
2018-11-21 19:05 ` [PATCH 4.19 09/42] tcp: Fix SOF_TIMESTAMPING_RX_HARDWARE to use the latest timestamp during TCP coalescing Greg Kroah-Hartman
2018-11-21 19:05 ` [PATCH 4.19 10/42] tg3: Add PHY reset for 5717/5719/5720 in change ring and flow control paths Greg Kroah-Hartman
2018-11-21 19:05 ` [PATCH 4.19 11/42] tipc: dont assume linear buffer when reading ancillary data Greg Kroah-Hartman
2018-11-21 19:05 ` [PATCH 4.19 12/42] tipc: fix lockdep warning when reinitilaizing sockets Greg Kroah-Hartman
2018-11-21 19:05 ` [PATCH 4.19 13/42] tuntap: fix multiqueue rx Greg Kroah-Hartman
2018-11-21 19:05 ` [PATCH 4.19 14/42] net: systemport: Protect stop from timeout Greg Kroah-Hartman
2018-11-21 19:05 ` [PATCH 4.19 15/42] net/sched: act_pedit: fix memory leak when IDR allocation fails Greg Kroah-Hartman
2018-11-21 19:05 ` [PATCH 4.19 16/42] net: sched: cls_flower: validate nested enc_opts_policy to avoid warning Greg Kroah-Hartman
2018-11-21 19:05 ` [PATCH 4.19 17/42] tipc: fix link re-establish failure Greg Kroah-Hartman
2018-11-21 19:05 ` [PATCH 4.19 18/42] net/mlx5e: Dont match on vlan non-existence if ethertype is wildcarded Greg Kroah-Hartman
2018-11-21 19:05 ` [PATCH 4.19 19/42] net/mlx5e: Claim TC hw offloads support only under a proper build config Greg Kroah-Hartman
2018-11-21 19:05 ` [PATCH 4.19 20/42] net/mlx5e: Adjust to max number of channles when re-attaching Greg Kroah-Hartman
2018-11-21 19:05 ` [PATCH 4.19 21/42] net/mlx5e: RX, verify received packet size in Linear Striding RQ Greg Kroah-Hartman
2018-11-21 19:05 ` [PATCH 4.19 22/42] Revert "sctp: remove sctp_transport_pmtu_check" Greg Kroah-Hartman
2018-11-21 19:05 ` [PATCH 4.19 23/42] net/mlx5e: Always use the match level enum when parsing TC rule match Greg Kroah-Hartman
2018-11-21 19:06 ` [PATCH 4.19 24/42] net/mlx5e: Fix selftest for small MTUs Greg Kroah-Hartman
2018-11-21 19:06 ` [PATCH 4.19 25/42] net/mlx5e: Removed unnecessary warnings in FEC caps query Greg Kroah-Hartman
2018-11-21 19:06 ` Greg Kroah-Hartman [this message]
2018-11-21 19:06 ` [PATCH 4.19 27/42] l2tp: fix a sock refcnt leak in l2tp_tunnel_register Greg Kroah-Hartman
2018-11-21 19:06 ` [PATCH 4.19 28/42] net/mlx5: IPSec, Fix the SA context hash key Greg Kroah-Hartman
2018-11-21 19:06 ` [PATCH 4.19 29/42] net/mlx5e: IPoIB, Reset QP after channels are closed Greg Kroah-Hartman
2018-11-21 19:06 ` [PATCH 4.19 30/42] net: dsa: mv88e6xxx: Fix clearing of stats counters Greg Kroah-Hartman
2018-11-21 19:06 ` [PATCH 4.19 31/42] net: phy: realtek: fix RTL8201F sysfs name Greg Kroah-Hartman
2018-11-21 19:06 ` [PATCH 4.19 32/42] sctp: define SCTP_SS_DEFAULT for Stream schedulers Greg Kroah-Hartman
2018-11-21 19:06 ` [PATCH 4.19 33/42] net: qualcomm: rmnet: Fix incorrect assignment of real_dev Greg Kroah-Hartman
2018-11-21 19:06 ` [PATCH 4.19 34/42] net: dsa: microchip: initialize mutex before use Greg Kroah-Hartman
2018-11-21 19:06 ` [PATCH 4.19 35/42] sctp: fix strchange_flags name for Stream Change Event Greg Kroah-Hartman
2018-11-21 19:06 ` [PATCH 4.19 36/42] net: phy: mdio-gpio: Fix working over slow can_sleep GPIOs Greg Kroah-Hartman
2018-11-21 19:06 ` [PATCH 4.19 37/42] sctp: not increase streams incnt before sending addstrm_in request Greg Kroah-Hartman
2018-11-21 19:06 ` [PATCH 4.19 38/42] mlxsw: spectrum: Fix IP2ME CPU policer configuration Greg Kroah-Hartman
2018-11-21 19:06 ` [PATCH 4.19 39/42] net: smsc95xx: Fix MTU range Greg Kroah-Hartman
2018-11-21 19:06 ` [PATCH 4.19 40/42] rxrpc: Fix lockup due to no error backoff after ack transmit error Greg Kroah-Hartman
2018-11-21 19:06 ` [PATCH 4.19 41/42] usbnet: smsc95xx: disable carrier check while suspending Greg Kroah-Hartman
2018-11-21 19:06 ` [PATCH 4.19 42/42] Revert "x86/speculation: Enable cross-hyperthread spectre v2 STIBP mitigation" Greg Kroah-Hartman
2018-11-22  5:26 ` [PATCH 4.19 00/42] 4.19.4-stable review kernelci.org bot
2018-11-22  7:15 ` Harsh Shandilya
2018-11-22  8:24   ` Greg Kroah-Hartman
2018-11-22 16:36 ` Guenter Roeck
2018-11-23  7:16   ` Greg Kroah-Hartman
2018-11-22 20:53 ` Thomas Voegtle
2018-11-22 22:01   ` Thomas Voegtle
2018-11-22 22:30     ` Guenter Roeck
2018-11-23  6:52       ` Greg Kroah-Hartman
2018-11-23  7:45       ` Greg Kroah-Hartman
2018-11-23 12:06         ` Guenter Roeck
2018-11-23  6:51   ` Greg Kroah-Hartman
2018-11-23 15:40     ` Thomas Voegtle
2018-11-23 15:44       ` David Laight
2018-11-23  8:13 ` Naresh Kamboju

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=20181121183149.133109771@linuxfoundation.org \
    --to=gregkh@linuxfoundation.org \
    --cc=davem@davemloft.net \
    --cc=edumazet@google.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=soukjin.bae@samsung.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).