public inbox for linux-kernel@vger.kernel.org
 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, CAI Qian <caiqian@redhat.com>,
	Patrick McHardy <kaber@trash.net>,
	Florian Westphal <fw@strlen.de>,
	Pablo Neira Ayuso <pablo@netfilter.org>
Subject: [ 18/73] netfilter: nf_nat: fix race when unloading protocol modules
Date: Thu,  9 May 2013 15:31:41 -0700	[thread overview]
Message-ID: <20130509222759.933753547@linuxfoundation.org> (raw)
In-Reply-To: <20130509222757.917088509@linuxfoundation.org>

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

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

From: Florian Westphal <fw@strlen.de>

commit c2d421e171868586939c328dfb91bab840fe4c49 upstream.

following oops was reported:
RIP: 0010:[<ffffffffa03227f2>]  [<ffffffffa03227f2>] nf_nat_cleanup_conntrack+0x42/0x70 [nf_nat]
RSP: 0018:ffff880202c63d40  EFLAGS: 00010246
RAX: 0000000000000000 RBX: ffff8801ac7bec28 RCX: ffff8801d0eedbe0
RDX: dead000000200200 RSI: 0000000000000011 RDI: ffffffffa03265b8
[..]
Call Trace:
 [..]
 [<ffffffffa02febed>] destroy_conntrack+0xbd/0x110 [nf_conntrack]

Happens when a conntrack timeout expires right after first part
of the nat cleanup has completed (bysrc hash removal), but before
part 2 has completed (re-initialization of nat area).

[ destroy callback tries to delete bysrc again ]

Patrick suggested to just remove the affected conntracks -- the
connections won't work properly anyway without nat transformation.

So, lets do that.

Reported-by: CAI Qian <caiqian@redhat.com>
Cc: Patrick McHardy <kaber@trash.net>
Signed-off-by: Florian Westphal <fw@strlen.de>
Acked-by: Patrick McHardy <kaber@trash.net>
Signed-off-by: Pablo Neira Ayuso <pablo@netfilter.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

---
 net/netfilter/nf_nat_core.c |   40 +++++++---------------------------------
 1 file changed, 7 insertions(+), 33 deletions(-)

--- a/net/netfilter/nf_nat_core.c
+++ b/net/netfilter/nf_nat_core.c
@@ -468,33 +468,22 @@ EXPORT_SYMBOL_GPL(nf_nat_packet);
 struct nf_nat_proto_clean {
 	u8	l3proto;
 	u8	l4proto;
-	bool	hash;
 };
 
-/* Clear NAT section of all conntracks, in case we're loaded again. */
-static int nf_nat_proto_clean(struct nf_conn *i, void *data)
+/* kill conntracks with affected NAT section */
+static int nf_nat_proto_remove(struct nf_conn *i, void *data)
 {
 	const struct nf_nat_proto_clean *clean = data;
 	struct nf_conn_nat *nat = nfct_nat(i);
 
 	if (!nat)
 		return 0;
-	if (!(i->status & IPS_SRC_NAT_DONE))
-		return 0;
+
 	if ((clean->l3proto && nf_ct_l3num(i) != clean->l3proto) ||
 	    (clean->l4proto && nf_ct_protonum(i) != clean->l4proto))
 		return 0;
 
-	if (clean->hash) {
-		spin_lock_bh(&nf_nat_lock);
-		hlist_del_rcu(&nat->bysource);
-		spin_unlock_bh(&nf_nat_lock);
-	} else {
-		memset(nat, 0, sizeof(*nat));
-		i->status &= ~(IPS_NAT_MASK | IPS_NAT_DONE_MASK |
-			       IPS_SEQ_ADJUST);
-	}
-	return 0;
+	return i->status & IPS_NAT_MASK ? 1 : 0;
 }
 
 static void nf_nat_l4proto_clean(u8 l3proto, u8 l4proto)
@@ -506,16 +495,8 @@ static void nf_nat_l4proto_clean(u8 l3pr
 	struct net *net;
 
 	rtnl_lock();
-	/* Step 1 - remove from bysource hash */
-	clean.hash = true;
 	for_each_net(net)
-		nf_ct_iterate_cleanup(net, nf_nat_proto_clean, &clean);
-	synchronize_rcu();
-
-	/* Step 2 - clean NAT section */
-	clean.hash = false;
-	for_each_net(net)
-		nf_ct_iterate_cleanup(net, nf_nat_proto_clean, &clean);
+		nf_ct_iterate_cleanup(net, nf_nat_proto_remove, &clean);
 	rtnl_unlock();
 }
 
@@ -527,16 +508,9 @@ static void nf_nat_l3proto_clean(u8 l3pr
 	struct net *net;
 
 	rtnl_lock();
-	/* Step 1 - remove from bysource hash */
-	clean.hash = true;
-	for_each_net(net)
-		nf_ct_iterate_cleanup(net, nf_nat_proto_clean, &clean);
-	synchronize_rcu();
 
-	/* Step 2 - clean NAT section */
-	clean.hash = false;
 	for_each_net(net)
-		nf_ct_iterate_cleanup(net, nf_nat_proto_clean, &clean);
+		nf_ct_iterate_cleanup(net, nf_nat_proto_remove, &clean);
 	rtnl_unlock();
 }
 
@@ -774,7 +748,7 @@ static void __net_exit nf_nat_net_exit(s
 {
 	struct nf_nat_proto_clean clean = {};
 
-	nf_ct_iterate_cleanup(net, &nf_nat_proto_clean, &clean);
+	nf_ct_iterate_cleanup(net, &nf_nat_proto_remove, &clean);
 	synchronize_rcu();
 	nf_ct_free_hashtable(net->ct.nat_bysource, net->ct.nat_htable_size);
 }



  parent reply	other threads:[~2013-05-09 22:32 UTC|newest]

Thread overview: 82+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-05-09 22:31 [ 00/73] 3.8.13-stable review Greg Kroah-Hartman
2013-05-09 22:31 ` [ 01/73] xen/arm: actually pass a non-NULL percpu pointer to request_percpu_irq Greg Kroah-Hartman
2013-05-09 22:31 ` [ 02/73] powerpc: Emulate non privileged DSCR read and write Greg Kroah-Hartman
2013-05-09 22:31 ` [ 03/73] powerpc: fix numa distance for form0 device tree Greg Kroah-Hartman
2013-05-09 22:31 ` [ 04/73] pwm: spear: Fix checking return value of clk_enable() and clk_prepare() Greg Kroah-Hartman
2013-05-09 22:31 ` [ 05/73] autofs - remove autofs dentry mount check Greg Kroah-Hartman
2013-05-09 22:31 ` [ 06/73] hugetlbfs: fix mmap failure in unaligned size request Greg Kroah-Hartman
2013-05-09 22:31 ` [ 07/73] iommu/amd: Properly initialize irq-table lock Greg Kroah-Hartman
2013-05-09 22:31 ` [ 08/73] net/eth/ibmveth: Fixup retrieval of MAC address Greg Kroah-Hartman
2013-05-09 22:31 ` [ 09/73] perf/x86/intel: Add support for IvyBridge model 58 Uncore Greg Kroah-Hartman
2013-05-09 22:31 ` [ 10/73] perf/x86/intel: Fix unintended variable name reuse Greg Kroah-Hartman
2013-05-09 22:31 ` [ 11/73] perf/x86/intel/lbr: Fix LBR filter Greg Kroah-Hartman
2013-05-09 22:31 ` [ 12/73] perf/x86/intel/lbr: Demand proper privileges for PERF_SAMPLE_BRANCH_KERNEL Greg Kroah-Hartman
2013-05-09 22:31 ` [ 13/73] PCI/PM: Clear state_saved during suspend Greg Kroah-Hartman
2013-05-09 22:31 ` [ 14/73] e1000e: fix runtime power management transitions Greg Kroah-Hartman
2013-05-09 22:31 ` [ 15/73] e1000e: fix accessing to suspended device Greg Kroah-Hartman
2013-05-09 22:31 ` [ 16/73] xhci: Dont warn on empty ring for suspended devices Greg Kroah-Hartman
2013-05-09 22:31 ` [ 17/73] ipvs: ip_vs_sip_fill_param() BUG: bad check of return value Greg Kroah-Hartman
2013-05-09 22:31 ` Greg Kroah-Hartman [this message]
2013-05-09 22:31 ` [ 19/73] netfilter: ipset: list:set: fix reference counter update Greg Kroah-Hartman
2013-05-09 22:31 ` [ 20/73] netfilter: nf_ct_sip: dont drop packets with offsets pointing outside the packet Greg Kroah-Hartman
2013-05-09 22:31 ` [ 21/73] netfilter: ipset: "Directory not empty" error message Greg Kroah-Hartman
2013-05-09 22:31 ` [ 22/73] netfilter: nf_ct_helper: dont discard helper if it is actually the same Greg Kroah-Hartman
2013-05-09 22:31 ` [ 23/73] netfilter: ctnetlink: dont permit ct creation with random tuple Greg Kroah-Hartman
2013-05-09 22:31 ` [ 24/73] netfilter: xt_rpfilter: skip locally generated broadcast/multicast, too Greg Kroah-Hartman
2013-05-09 22:31 ` [ 25/73] netfilter: ip6t_NPT: Fix translation for non-multiple of 32 prefix lengths Greg Kroah-Hartman
2013-05-09 22:31 ` [ 26/73] ext4: add check for inodes_count overflow in new resize ioctl Greg Kroah-Hartman
2013-05-09 22:31 ` [ 27/73] r8169: fix 8168evl frame padding Greg Kroah-Hartman
2013-05-09 22:31 ` [ 28/73] RDMA/cxgb4: Fix SQ allocation when on-chip SQ is disabled Greg Kroah-Hartman
2013-05-09 22:31 ` [ 29/73] arm64: Ignore the write ESR flag on cache maintenance faults Greg Kroah-Hartman
2013-05-09 22:31 ` [ 30/73] block: fix max discard sectors limit Greg Kroah-Hartman
2013-05-09 22:31 ` [ 31/73] drm/cirrus: deal with bo reserve fail in dirty update path Greg Kroah-Hartman
2013-05-09 22:31 ` [ 32/73] drm/mgag200: " Greg Kroah-Hartman
2013-05-09 22:31 ` [ 33/73] drm/gma500: fix backlight hotkeys behaviour on netbooks Greg Kroah-Hartman
2013-05-09 22:31 ` [ 34/73] drm/prime: keep a reference from the handle to exported dma-buf (v6) Greg Kroah-Hartman
2013-05-09 22:31 ` [ 35/73] drm/ast: deal with bo reserve fail in dirty update path Greg Kroah-Hartman
2013-05-09 22:31 ` [ 36/73] drm/i915: Fix detection of base of stolen memory Greg Kroah-Hartman
2013-05-09 22:32 ` [ 37/73] drm/i915: Fix sdvo connector get_hw_state function Greg Kroah-Hartman
2013-05-09 22:32 ` [ 38/73] drm/i915: Add no-lvds quirk for Fujitsu Esprimo Q900 Greg Kroah-Hartman
2013-05-09 22:32 ` [ 39/73] drm/i915: Fix SDVO connector and encoder get_hw_state functions Greg Kroah-Hartman
2013-05-09 22:32 ` [ 40/73] drm/i915: Workaround incoherence between fences and LLC across multiple CPUs Greg Kroah-Hartman
2013-05-09 22:32 ` [ 41/73] drm/i915: Use MLC (l3$) for context objects Greg Kroah-Hartman
2013-05-09 22:32 ` [ 42/73] drm/i915: set CPT FDI RX polarity bits based on VBT Greg Kroah-Hartman
2013-05-09 22:32 ` [ 43/73] drm/i915: ensure single initialization and cleanup of backlight device Greg Kroah-Hartman
2013-05-09 22:32 ` [ 44/73] drm/i915: Fixup Oops in the pipe config computation Greg Kroah-Hartman
2013-05-09 22:32 ` [ 45/73] drm/i915: Fall back to bit banging mode for DVO transmitter detection Greg Kroah-Hartman
2013-05-09 22:32 ` [ 46/73] drm/radeon: dont use get_engine_clock() on APUs Greg Kroah-Hartman
2013-05-09 22:32 ` [ 47/73] drm/radeon: use frac fb div on RS780/RS880 Greg Kroah-Hartman
2013-05-09 22:32 ` [ 48/73] drm/radeon: fix typo in rv515_mc_resume() Greg Kroah-Hartman
2013-05-09 22:32 ` [ 49/73] drm/radeon/dce6: add missing display reg for tiling setup Greg Kroah-Hartman
2013-05-09 22:32 ` [ 50/73] drm/radeon: update wait_for_vblank for r5xx-r7xx Greg Kroah-Hartman
2013-05-09 22:32 ` [ 51/73] drm/radeon: update wait_for_vblank for evergreen+ Greg Kroah-Hartman
2013-05-09 22:32 ` [ 52/73] drm/radeon: properly lock disp in mc_stop/resume " Greg Kroah-Hartman
2013-05-09 22:32 ` [ 53/73] drm/radeon: properly lock disp in mc_stop/resume for r5xx-r7xx Greg Kroah-Hartman
2013-05-09 22:32 ` [ 54/73] drm/radeon: update wait_for_vblank for r1xx-r4xx Greg Kroah-Hartman
2013-05-09 22:32 ` [ 55/73] drm/radeon: disable the crtcs in mc_stop (evergreen+) (v2) Greg Kroah-Hartman
2013-05-09 22:32 ` [ 56/73] drm/radeon: add some new SI PCI ids Greg Kroah-Hartman
2013-05-09 22:32 ` [ 57/73] drm/radeon/evergreen+: dont enable HPD interrupts on eDP/LVDS Greg Kroah-Hartman
2013-05-09 22:32 ` [ 58/73] drm/radeon: cleanup properly if mmio mapping fails Greg Kroah-Hartman
2013-05-09 22:32 ` [ 59/73] drm/radeon: fix hdmi mode enable on RS600/RS690/RS740 Greg Kroah-Hartman
2013-05-09 22:32 ` [ 60/73] drm/radeon: fix typo in si_select_se_sh() Greg Kroah-Hartman
2013-05-09 22:32 ` [ 61/73] drm/radeon: Always flush the VM Greg Kroah-Hartman
2013-05-09 22:32 ` [ 62/73] drm/radeon: disable the crtcs in mc_stop (r5xx-r7xx) (v2) Greg Kroah-Hartman
2013-05-09 22:32 ` [ 63/73] drm/radeon: fix endian bugs in atom_allocate_fb_scratch() Greg Kroah-Hartman
2013-05-09 22:32 ` [ 64/73] drm/radeon: fix possible segfault when parsing pm tables Greg Kroah-Hartman
2013-05-09 22:32 ` [ 65/73] drm/radeon: add new richland pci ids Greg Kroah-Hartman
2013-05-09 22:32 ` [ 66/73] drm/radeon: fix handling of v6 power tables Greg Kroah-Hartman
2013-05-09 22:32 ` [ 67/73] tracing: Fix ftrace_dump() Greg Kroah-Hartman
2013-05-09 22:32 ` [ 68/73] Btrfs: compare relevant parts of delayed tree refs Greg Kroah-Hartman
2013-05-09 22:32 ` [ 69/73] Btrfs: fix extent logging with O_DIRECT into prealloc Greg Kroah-Hartman
2013-05-09 22:32 ` [ 70/73] EDAC: Dont give write permission to read-only files Greg Kroah-Hartman
2013-05-09 22:32 ` [ 71/73] NFSv4.x: Fix handling of partially delegated locks Greg Kroah-Hartman
2013-05-09 22:32 ` [ 72/73] kernel/audit_tree.c: tree will leak memory when failure occurs in audit_trim_trees() Greg Kroah-Hartman
2013-05-09 22:32 ` [ 73/73] x86/mm: account for PGDIR_SIZE alignment Greg Kroah-Hartman
2013-05-10 10:54 ` [ 00/73] 3.8.13-stable review Holger Hoffstaette
2013-05-15  0:07   ` r8169 on 3.8.13, 3.9.2, 3.10-rc1, was " Ken Moffat
2013-05-15  6:14     ` Francois Romieu
2013-05-15 17:09       ` Ken Moffat
2013-05-15 20:39       ` David Miller
2013-05-15 23:15         ` David Miller
2013-05-10 15:24 ` Shuah Khan
2013-05-11  6:03 ` Satoru Takeuchi

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=20130509222759.933753547@linuxfoundation.org \
    --to=gregkh@linuxfoundation.org \
    --cc=caiqian@redhat.com \
    --cc=fw@strlen.de \
    --cc=kaber@trash.net \
    --cc=linux-kernel@vger.kernel.org \
    --cc=pablo@netfilter.org \
    --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