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, Slava Shwartsman <slavash@mellanox.com>,
	Herbert Xu <herbert@gondor.apana.org.au>,
	Eric Dumazet <edumazet@google.com>,
	"David S. Miller" <davem@davemloft.net>
Subject: [PATCH 4.9 34/59] gro: Disable frag0 optimization on IPv6 ext headers
Date: Fri, 13 Jan 2017 13:01:41 +0100	[thread overview]
Message-ID: <20170113113840.664166132@linuxfoundation.org> (raw)
In-Reply-To: <20170113113839.364876751@linuxfoundation.org>

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

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

From: Herbert Xu <herbert@gondor.apana.org.au>


[ Upstream commit 57ea52a865144aedbcd619ee0081155e658b6f7d ]

The GRO fast path caches the frag0 address.  This address becomes
invalid if frag0 is modified by pskb_may_pull or its variants.
So whenever that happens we must disable the frag0 optimization.

This is usually done through the combination of gro_header_hard
and gro_header_slow, however, the IPv6 extension header path did
the pulling directly and would continue to use the GRO fast path
incorrectly.

This patch fixes it by disabling the fast path when we enter the
IPv6 extension header path.

Fixes: 78a478d0efd9 ("gro: Inline skb_gro_header and cache frag0 virtual address")
Reported-by: Slava Shwartsman <slavash@mellanox.com>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
 include/linux/netdevice.h |    9 +++++++--
 net/ipv6/ip6_offload.c    |    1 +
 2 files changed, 8 insertions(+), 2 deletions(-)

--- a/include/linux/netdevice.h
+++ b/include/linux/netdevice.h
@@ -2502,14 +2502,19 @@ static inline int skb_gro_header_hard(st
 	return NAPI_GRO_CB(skb)->frag0_len < hlen;
 }
 
+static inline void skb_gro_frag0_invalidate(struct sk_buff *skb)
+{
+	NAPI_GRO_CB(skb)->frag0 = NULL;
+	NAPI_GRO_CB(skb)->frag0_len = 0;
+}
+
 static inline void *skb_gro_header_slow(struct sk_buff *skb, unsigned int hlen,
 					unsigned int offset)
 {
 	if (!pskb_may_pull(skb, hlen))
 		return NULL;
 
-	NAPI_GRO_CB(skb)->frag0 = NULL;
-	NAPI_GRO_CB(skb)->frag0_len = 0;
+	skb_gro_frag0_invalidate(skb);
 	return skb->data + offset;
 }
 
--- a/net/ipv6/ip6_offload.c
+++ b/net/ipv6/ip6_offload.c
@@ -191,6 +191,7 @@ static struct sk_buff **ipv6_gro_receive
 	ops = rcu_dereference(inet6_offloads[proto]);
 	if (!ops || !ops->callbacks.gro_receive) {
 		__pskb_pull(skb, skb_gro_offset(skb));
+		skb_gro_frag0_invalidate(skb);
 		proto = ipv6_gso_pull_exthdrs(skb, proto);
 		skb_gro_pull(skb, -skb_transport_offset(skb));
 		skb_reset_transport_header(skb);



  parent reply	other threads:[~2017-01-13 12:03 UTC|newest]

Thread overview: 61+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
     [not found] <CGME20170113120315epcas3p230e25f6a27069614fecfb95292c2ba78@epcas3p2.samsung.com>
2017-01-13 12:01 ` [PATCH 4.9 00/59] 4.9.4-stable review Greg Kroah-Hartman
2017-01-13 12:01   ` [PATCH 4.9 01/59] net: vrf: Fix NAT within a VRF Greg Kroah-Hartman
2017-01-13 12:01   ` [PATCH 4.9 02/59] net: vrf: Drop conntrack data after pass through VRF device on Tx Greg Kroah-Hartman
2017-01-13 12:01   ` [PATCH 4.9 03/59] sctp: sctp_transport_lookup_process should rcu_read_unlock when transport is null Greg Kroah-Hartman
2017-01-13 12:01   ` [PATCH 4.9 04/59] inet: fix IP(V6)_RECVORIGDSTADDR for udp sockets Greg Kroah-Hartman
2017-01-13 12:01   ` [PATCH 4.9 05/59] ipv6: handle -EFAULT from skb_copy_bits Greg Kroah-Hartman
2017-01-13 12:01   ` [PATCH 4.9 06/59] net, sched: fix soft lockup in tc_classify Greg Kroah-Hartman
2017-01-13 12:01   ` [PATCH 4.9 07/59] net: stmmac: Fix race between stmmac_drv_probe and stmmac_open Greg Kroah-Hartman
2017-01-13 12:01   ` [PATCH 4.9 08/59] net/sched: cls_flower: Fix missing addr_type in classify Greg Kroah-Hartman
2017-01-13 12:01   ` [PATCH 4.9 09/59] net/mlx5: Check FW limitations on log_max_qp before setting it Greg Kroah-Hartman
2017-01-13 12:01   ` [PATCH 4.9 10/59] net/mlx5: Cancel recovery work in remove flow Greg Kroah-Hartman
2017-01-13 12:01   ` [PATCH 4.9 11/59] net/mlx5: Avoid shadowing numa_node Greg Kroah-Hartman
2017-01-13 12:01   ` [PATCH 4.9 12/59] net/mlx5: Mask destination mac value in ethtool steering rules Greg Kroah-Hartman
2017-01-13 12:01   ` [PATCH 4.9 13/59] net/mlx5: Prevent setting multicast macs for VFs Greg Kroah-Hartman
2017-01-13 12:01   ` [PATCH 4.9 14/59] net/mlx5e: Dont sync netdev state when not registered Greg Kroah-Hartman
2017-01-13 12:01   ` [PATCH 4.9 15/59] net/mlx5e: Disable netdev after close Greg Kroah-Hartman
2017-01-13 12:01   ` [PATCH 4.9 16/59] rtnl: stats - add missing netlink message size checks Greg Kroah-Hartman
2017-01-13 12:01   ` [PATCH 4.9 17/59] net: fix incorrect original ingress device index in PKTINFO Greg Kroah-Hartman
2017-01-13 12:01   ` [PATCH 4.9 18/59] net: ipv4: dst for local input routes should use l3mdev if relevant Greg Kroah-Hartman
2017-01-13 12:01   ` [PATCH 4.9 19/59] drop_monitor: add missing call to genlmsg_end Greg Kroah-Hartman
2017-01-13 12:01   ` [PATCH 4.9 20/59] drop_monitor: consider inserted data in genlmsg_end Greg Kroah-Hartman
2017-01-13 12:01   ` [PATCH 4.9 21/59] flow_dissector: Update pptp handling to avoid null pointer deref Greg Kroah-Hartman
2017-01-13 12:01   ` [PATCH 4.9 22/59] igmp: Make igmp group member RFC 3376 compliant Greg Kroah-Hartman
2017-01-13 12:01   ` [PATCH 4.9 23/59] ipv4: Do not allow MAIN to be alias for new LOCAL w/ custom rules Greg Kroah-Hartman
2017-01-13 12:01   ` [PATCH 4.9 24/59] net: vrf: Add missing Rx counters Greg Kroah-Hartman
2017-01-13 12:01   ` [PATCH 4.9 25/59] bpf: change back to orig prog on too many passes Greg Kroah-Hartman
2017-01-13 12:01   ` [PATCH 4.9 26/59] net: dsa: bcm_sf2: Do not clobber b53_switch_ops Greg Kroah-Hartman
2017-01-13 12:01   ` [PATCH 4.9 27/59] net: dsa: bcm_sf2: Utilize nested MDIO read/write Greg Kroah-Hartman
2017-01-13 12:01   ` [PATCH 4.9 28/59] r8152: split rtl8152_suspend function Greg Kroah-Hartman
2017-01-13 12:01   ` [PATCH 4.9 29/59] r8152: fix rx issue for runtime suspend Greg Kroah-Hartman
2017-01-13 12:01   ` [PATCH 4.9 30/59] net: dsa: Ensure validity of dst->ds[0] Greg Kroah-Hartman
2017-01-13 12:01   ` [PATCH 4.9 31/59] net: add the AF_QIPCRTR entries to family name tables Greg Kroah-Hartman
2017-01-13 12:01   ` [PATCH 4.9 32/59] gro: Enter slow-path if there is no tailroom Greg Kroah-Hartman
2017-01-13 12:01   ` [PATCH 4.9 33/59] gro: use min_t() in skb_gro_reset_offset() Greg Kroah-Hartman
2017-01-13 12:01   ` Greg Kroah-Hartman [this message]
2017-01-13 12:01   ` [PATCH 4.9 35/59] net/mlx5e: Remove WARN_ONCE from adaptive moderation code Greg Kroah-Hartman
2017-01-13 12:01   ` [PATCH 4.9 36/59] net: ipv4: Fix multipath selection with vrf Greg Kroah-Hartman
2017-01-13 12:01   ` [PATCH 4.9 37/59] net: vrf: do not allow table id 0 Greg Kroah-Hartman
2017-01-13 12:01   ` [PATCH 4.9 38/59] HID: hid-cypress: validate length of report Greg Kroah-Hartman
2017-01-13 12:01   ` [PATCH 4.9 39/59] ALSA: firewire-tascam: Fix to handle error from initialization of stream data Greg Kroah-Hartman
2017-01-13 12:01   ` [PATCH 4.9 40/59] powerpc: Fix build warning on 32-bit PPC Greg Kroah-Hartman
2017-01-13 12:01   ` [PATCH 4.9 41/59] tools/virtio: fix READ_ONCE() Greg Kroah-Hartman
2017-01-13 12:01   ` [PATCH 4.9 42/59] arm64: dts: mt8173: Fix auxadc node Greg Kroah-Hartman
2017-01-13 12:01   ` [PATCH 4.9 43/59] ARM64: dts: bcm2837-rpi-3-b: remove incorrect pwr LED Greg Kroah-Hartman
2017-01-13 12:01   ` [PATCH 4.9 45/59] ARM: pxa: fix pxa25x interrupt init Greg Kroah-Hartman
2017-01-13 12:01   ` [PATCH 4.9 46/59] ARM: zynq: Reserve correct amount of non-DMA RAM Greg Kroah-Hartman
2017-01-13 12:01   ` [PATCH 4.9 47/59] ARM: qcom_defconfig: Fix MDM9515 LCC and GCC config Greg Kroah-Hartman
2017-01-13 12:01   ` [PATCH 4.9 48/59] svcrdma: Clear xpt_bc_xps in xprt_setup_rdma_bc() error exit arm Greg Kroah-Hartman
2017-01-13 12:01   ` [PATCH 4.9 49/59] bus: arm-ccn: Prevent hotplug callback leak Greg Kroah-Hartman
2017-01-13 12:01   ` [PATCH 4.9 50/59] ARM: OMAP5: Fix mpuss_early_init Greg Kroah-Hartman
2017-01-13 12:01   ` [PATCH 4.9 51/59] ARM: OMAP5: Fix build for PM code Greg Kroah-Hartman
2017-01-13 12:01   ` [PATCH 4.9 52/59] ARM: OMAP4+: Fix bad fallthrough for cpuidle Greg Kroah-Hartman
2017-01-13 12:02   ` [PATCH 4.9 53/59] ARM: 8631/1: clkdev: Detect errors in clk_hw_register_clkdev() for mass registration Greg Kroah-Hartman
2017-01-13 12:02   ` [PATCH 4.9 54/59] ARM: omap2+: am437x: rollback to use omap3_gptimer_timer_init() Greg Kroah-Hartman
2017-01-13 12:02   ` [PATCH 4.9 56/59] ALSA: usb-audio: Add a quirk for Plantronics BT600 Greg Kroah-Hartman
2017-01-13 12:02   ` [PATCH 4.9 58/59] rtlwifi: Fix enter/exit power_save Greg Kroah-Hartman
2017-01-13 12:02   ` [PATCH 4.9 59/59] rtlwifi: rtl_usb: Fix missing entry in USB drivers private data Greg Kroah-Hartman
2017-01-13 20:20   ` [PATCH 4.9 00/59] 4.9.4-stable review Guenter Roeck
2017-01-14  7:23     ` Greg Kroah-Hartman
2017-01-13 21:58   ` Shuah Khan
2017-01-14  7:24     ` Greg Kroah-Hartman

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=20170113113840.664166132@linuxfoundation.org \
    --to=gregkh@linuxfoundation.org \
    --cc=davem@davemloft.net \
    --cc=edumazet@google.com \
    --cc=herbert@gondor.apana.org.au \
    --cc=linux-kernel@vger.kernel.org \
    --cc=slavash@mellanox.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).