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>,
Juha-Matti Tilli <juha-matti.tilli@iki.fi>,
Yuchung Cheng <ycheng@google.com>,
Soheil Hassas Yeganeh <soheil@google.com>,
"David S. Miller" <davem@davemloft.net>,
Mao Wenan <maowenan@huawei.com>
Subject: [PATCH 4.4 23/27] tcp: free batches of packets in tcp_prune_ofo_queue()
Date: Thu, 11 Oct 2018 17:35:10 +0200 [thread overview]
Message-ID: <20181011152535.140663022@linuxfoundation.org> (raw)
In-Reply-To: <20181011152534.014964888@linuxfoundation.org>
4.4-stable review patch. If anyone has any objections, please let me know.
------------------
From: Eric Dumazet <edumazet@google.com>
[ Upstream commit 72cd43ba64fc172a443410ce01645895850844c8 ]
Juha-Matti Tilli reported that malicious peers could inject tiny
packets in out_of_order_queue, forcing very expensive calls
to tcp_collapse_ofo_queue() and tcp_prune_ofo_queue() for
every incoming packet. out_of_order_queue rb-tree can contain
thousands of nodes, iterating over all of them is not nice.
Before linux-4.9, we would have pruned all packets in ofo_queue
in one go, every XXXX packets. XXXX depends on sk_rcvbuf and skbs
truesize, but is about 7000 packets with tcp_rmem[2] default of 6 MB.
Since we plan to increase tcp_rmem[2] in the future to cope with
modern BDP, can not revert to the old behavior, without great pain.
Strategy taken in this patch is to purge ~12.5 % of the queue capacity.
Fixes: 36a6503fedda ("tcp: refine tcp_prune_ofo_queue() to not drop all packets")
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reported-by: Juha-Matti Tilli <juha-matti.tilli@iki.fi>
Acked-by: Yuchung Cheng <ycheng@google.com>
Acked-by: Soheil Hassas Yeganeh <soheil@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Mao Wenan <maowenan@huawei.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
net/ipv4/tcp_input.c | 16 +++++++++++-----
1 file changed, 11 insertions(+), 5 deletions(-)
--- a/net/ipv4/tcp_input.c
+++ b/net/ipv4/tcp_input.c
@@ -4899,27 +4899,33 @@ new_range:
/*
* Purge the out-of-order queue.
+ * Drop at least 12.5 % of sk_rcvbuf to avoid malicious attacks.
* Return true if queue was pruned.
*/
static bool tcp_prune_ofo_queue(struct sock *sk)
{
struct tcp_sock *tp = tcp_sk(sk);
struct rb_node *node, *prev;
+ int goal;
if (RB_EMPTY_ROOT(&tp->out_of_order_queue))
return false;
NET_INC_STATS_BH(sock_net(sk), LINUX_MIB_OFOPRUNED);
-
+ goal = sk->sk_rcvbuf >> 3;
node = &tp->ooo_last_skb->rbnode;
do {
prev = rb_prev(node);
rb_erase(node, &tp->out_of_order_queue);
+ goal -= rb_to_skb(node)->truesize;
__kfree_skb(rb_to_skb(node));
- sk_mem_reclaim(sk);
- if (atomic_read(&sk->sk_rmem_alloc) <= sk->sk_rcvbuf &&
- !tcp_under_memory_pressure(sk))
- break;
+ if (!prev || goal <= 0) {
+ sk_mem_reclaim(sk);
+ if (atomic_read(&sk->sk_rmem_alloc) <= sk->sk_rcvbuf &&
+ !tcp_under_memory_pressure(sk))
+ break;
+ goal = sk->sk_rcvbuf >> 3;
+ }
node = prev;
} while (node);
next prev parent reply other threads:[~2018-10-11 15:42 UTC|newest]
Thread overview: 35+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-10-11 15:34 [PATCH 4.4 00/27] 4.4.161-stable review Greg Kroah-Hartman
2018-10-11 15:34 ` [PATCH 4.4 01/27] mm/vmstat.c: skip NR_TLB_REMOTE_FLUSH* properly Greg Kroah-Hartman
2018-10-11 15:34 ` [PATCH 4.4 02/27] fbdev/omapfb: fix omapfb_memory_read infoleak Greg Kroah-Hartman
2018-10-11 15:34 ` [PATCH 4.4 03/27] x86/vdso: Fix asm constraints on vDSO syscall fallbacks Greg Kroah-Hartman
2018-10-11 15:34 ` [PATCH 4.4 04/27] x86/vdso: Fix vDSO syscall fallback asm constraint regression Greg Kroah-Hartman
2018-10-11 15:34 ` [PATCH 4.4 05/27] PCI: Reprogram bridge prefetch registers on resume Greg Kroah-Hartman
2018-10-11 15:34 ` [PATCH 4.4 06/27] mac80211: fix setting IEEE80211_KEY_FLAG_RX_MGMT for AP mode keys Greg Kroah-Hartman
2018-10-11 15:34 ` [PATCH 4.4 07/27] PM / core: Clear the direct_complete flag on errors Greg Kroah-Hartman
2018-10-11 15:34 ` [PATCH 4.4 08/27] dm cache: fix resize crash if user doesnt reload cache table Greg Kroah-Hartman
2018-10-11 15:34 ` [PATCH 4.4 09/27] xhci: Add missing CAS workaround for Intel Sunrise Point xHCI Greg Kroah-Hartman
2018-10-11 15:34 ` [PATCH 4.4 10/27] USB: serial: simple: add Motorola Tetra MTP6550 id Greg Kroah-Hartman
2018-10-11 15:34 ` [PATCH 4.4 11/27] of: unittest: Disable interrupt node tests for old world MAC systems Greg Kroah-Hartman
2018-10-11 15:34 ` [PATCH 4.4 12/27] ext4: add corruption check in ext4_xattr_set_entry() Greg Kroah-Hartman
2018-10-11 15:35 ` [PATCH 4.4 13/27] ext4: always verify the magic number in xattr blocks Greg Kroah-Hartman
2018-10-11 15:35 ` [PATCH 4.4 14/27] cgroup: Fix deadlock in cpu hotplug path Greg Kroah-Hartman
2018-10-11 15:35 ` [PATCH 4.4 15/27] ath10k: fix use-after-free in ath10k_wmi_cmd_send_nowait Greg Kroah-Hartman
2018-10-11 15:35 ` [PATCH 4.4 16/27] powerpc/fadump: Return error when fadump registration fails Greg Kroah-Hartman
2018-10-11 15:35 ` [PATCH 4.4 17/27] ARC: clone syscall to setp r25 as thread pointer Greg Kroah-Hartman
2018-10-11 15:35 ` [PATCH 4.4 18/27] ucma: fix a use-after-free in ucma_resolve_ip() Greg Kroah-Hartman
2018-10-11 15:35 ` [PATCH 4.4 19/27] ubifs: Check for name being NULL while mounting Greg Kroah-Hartman
2018-10-11 15:35 ` [PATCH 4.4 20/27] tcp: increment sk_drops for dropped rx packets Greg Kroah-Hartman
2018-10-11 15:35 ` [PATCH 4.4 21/27] tcp: use an RB tree for ooo receive queue Greg Kroah-Hartman
2018-10-11 15:35 ` [PATCH 4.4 22/27] tcp: fix a stale ooo_last_skb after a replace Greg Kroah-Hartman
2018-10-11 15:35 ` Greg Kroah-Hartman [this message]
2018-10-11 15:35 ` [PATCH 4.4 24/27] tcp: call tcp_drop() from tcp_data_queue_ofo() Greg Kroah-Hartman
2018-10-11 15:35 ` [PATCH 4.4 25/27] tcp: add tcp_ooo_try_coalesce() helper Greg Kroah-Hartman
2018-10-11 15:35 ` [PATCH 4.4 26/27] ath10k: fix scan crash due to incorrect length calculation Greg Kroah-Hartman
2018-10-11 15:35 ` [PATCH 4.4 27/27] ebtables: arpreply: Add the standard target sanity check Greg Kroah-Hartman
2018-10-11 22:43 ` [PATCH 4.4 00/27] 4.4.161-stable review Shuah Khan
2018-10-12 4:38 ` Naresh Kamboju
[not found] ` <5bbfcd13.1c69fb81.477c6.d7d9@mx.google.com>
2018-10-12 6:14 ` Greg Kroah-Hartman
2018-10-12 12:22 ` Guenter Roeck
2018-10-12 7:52 ` Jon Hunter
2018-10-12 15:42 ` Guenter Roeck
2018-10-12 17:07 ` Nathan Chancellor
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=20181011152535.140663022@linuxfoundation.org \
--to=gregkh@linuxfoundation.org \
--cc=davem@davemloft.net \
--cc=edumazet@google.com \
--cc=juha-matti.tilli@iki.fi \
--cc=linux-kernel@vger.kernel.org \
--cc=maowenan@huawei.com \
--cc=soheil@google.com \
--cc=stable@vger.kernel.org \
--cc=ycheng@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).