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, "Jason A. Donenfeld" <Jason@zx2c4.com>,
	Hannes Frederic Sowa <hannes@stressinduktion.org>,
	"David S. Miller" <davem@davemloft.net>
Subject: [PATCH 3.14 07/47] ip_tunnel: disable preemption when updating per-cpu tstats
Date: Wed, 20 Jan 2016 14:00:39 -0800	[thread overview]
Message-ID: <20160120215508.714083342@linuxfoundation.org> (raw)
In-Reply-To: <20160120215507.575738941@linuxfoundation.org>

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

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

From: "Jason A. Donenfeld" <Jason@zx2c4.com>

[ Upstream commit b4fe85f9c9146f60457e9512fb6055e69e6a7a65 ]

Drivers like vxlan use the recently introduced
udp_tunnel_xmit_skb/udp_tunnel6_xmit_skb APIs. udp_tunnel6_xmit_skb
makes use of ip6tunnel_xmit, and ip6tunnel_xmit, after sending the
packet, updates the struct stats using the usual
u64_stats_update_begin/end calls on this_cpu_ptr(dev->tstats).
udp_tunnel_xmit_skb makes use of iptunnel_xmit, which doesn't touch
tstats, so drivers like vxlan, immediately after, call
iptunnel_xmit_stats, which does the same thing - calls
u64_stats_update_begin/end on this_cpu_ptr(dev->tstats).

While vxlan is probably fine (I don't know?), calling a similar function
from, say, an unbound workqueue, on a fully preemptable kernel causes
real issues:

[  188.434537] BUG: using smp_processor_id() in preemptible [00000000] code: kworker/u8:0/6
[  188.435579] caller is debug_smp_processor_id+0x17/0x20
[  188.435583] CPU: 0 PID: 6 Comm: kworker/u8:0 Not tainted 4.2.6 #2
[  188.435607] Call Trace:
[  188.435611]  [<ffffffff8234e936>] dump_stack+0x4f/0x7b
[  188.435615]  [<ffffffff81915f3d>] check_preemption_disabled+0x19d/0x1c0
[  188.435619]  [<ffffffff81915f77>] debug_smp_processor_id+0x17/0x20

The solution would be to protect the whole
this_cpu_ptr(dev->tstats)/u64_stats_update_begin/end blocks with
disabling preemption and then reenabling it.

Signed-off-by: Jason A. Donenfeld <Jason@zx2c4.com>
Acked-by: Hannes Frederic Sowa <hannes@stressinduktion.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
 include/net/ip6_tunnel.h |    3 ++-
 include/net/ip_tunnels.h |    3 ++-
 2 files changed, 4 insertions(+), 2 deletions(-)

--- a/include/net/ip6_tunnel.h
+++ b/include/net/ip6_tunnel.h
@@ -79,11 +79,12 @@ static inline void ip6tunnel_xmit(struct
 	err = ip6_local_out(skb);
 
 	if (net_xmit_eval(err) == 0) {
-		struct pcpu_sw_netstats *tstats = this_cpu_ptr(dev->tstats);
+		struct pcpu_sw_netstats *tstats = get_cpu_ptr(dev->tstats);
 		u64_stats_update_begin(&tstats->syncp);
 		tstats->tx_bytes += pkt_len;
 		tstats->tx_packets++;
 		u64_stats_update_end(&tstats->syncp);
+		put_cpu_ptr(tstats);
 	} else {
 		stats->tx_errors++;
 		stats->tx_aborted_errors++;
--- a/include/net/ip_tunnels.h
+++ b/include/net/ip_tunnels.h
@@ -166,12 +166,13 @@ static inline void iptunnel_xmit_stats(i
 				       struct pcpu_sw_netstats __percpu *stats)
 {
 	if (err > 0) {
-		struct pcpu_sw_netstats *tstats = this_cpu_ptr(stats);
+		struct pcpu_sw_netstats *tstats = get_cpu_ptr(stats);
 
 		u64_stats_update_begin(&tstats->syncp);
 		tstats->tx_bytes += err;
 		tstats->tx_packets++;
 		u64_stats_update_end(&tstats->syncp);
+		put_cpu_ptr(tstats);
 	} else if (err < 0) {
 		err_stats->tx_errors++;
 		err_stats->tx_aborted_errors++;

  parent reply	other threads:[~2016-01-20 22:00 UTC|newest]

Thread overview: 48+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-01-20 22:00 [PATCH 3.14 00/47] 3.14.59-stable review Greg Kroah-Hartman
2016-01-20 22:00 ` [PATCH 3.14 01/47] unix: avoid use-after-free in ep_remove_wait_queue Greg Kroah-Hartman
2016-01-20 22:00 ` [PATCH 3.14 02/47] tools/net: Use include/uapi with __EXPORTED_HEADERS__ Greg Kroah-Hartman
2016-01-20 22:00 ` [PATCH 3.14 03/47] packet: do skb_probe_transport_header when we actually have data Greg Kroah-Hartman
2016-01-20 22:00 ` [PATCH 3.14 04/47] packet: always probe for transport header Greg Kroah-Hartman
2016-01-20 22:00 ` [PATCH 3.14 05/47] packet: infer protocol from ethernet header if unset Greg Kroah-Hartman
2016-01-20 22:00 ` [PATCH 3.14 06/47] sctp: translate host order to network order when setting a hmacid Greg Kroah-Hartman
2016-01-20 22:00 ` Greg Kroah-Hartman [this message]
2016-01-20 22:00 ` [PATCH 3.14 08/47] snmp: Remove duplicate OUTMCAST stat increment Greg Kroah-Hartman
2016-01-20 22:00 ` [PATCH 3.14 10/47] tcp: md5: fix lockdep annotation Greg Kroah-Hartman
2016-01-20 22:00 ` [PATCH 3.14 11/47] tcp: initialize tp->copied_seq in case of cross SYN connection Greg Kroah-Hartman
2016-01-20 22:00 ` [PATCH 3.14 12/47] net, scm: fix PaX detected msg_controllen overflow in scm_detach_fds Greg Kroah-Hartman
2016-01-20 22:00 ` [PATCH 3.14 13/47] net: ipmr: fix static mfc/dev leaks on table destruction Greg Kroah-Hartman
2016-01-20 22:00 ` [PATCH 3.14 14/47] net: ip6mr: " Greg Kroah-Hartman
2016-01-20 22:00 ` [PATCH 3.14 15/47] broadcom: fix PHY_ID_BCM5481 entry in the id table Greg Kroah-Hartman
2016-01-20 22:00 ` [PATCH 3.14 16/47] ipv6: distinguish frag queues by device for multicast and link-local packets Greg Kroah-Hartman
2016-01-20 22:00 ` [PATCH 3.14 17/47] ipv6: sctp: implement sctp_v6_destroy_sock() Greg Kroah-Hartman
2016-01-20 22:00 ` [PATCH 3.14 18/47] Btrfs: fix race leading to incorrect item deletion when dropping extents Greg Kroah-Hartman
2016-01-20 22:00 ` [PATCH 3.14 19/47] Btrfs: fix race leading to BUG_ON when running delalloc for nodatacow Greg Kroah-Hartman
2016-01-20 22:00 ` [PATCH 3.14 20/47] ext4: fix potential use after free in __ext4_journal_stop Greg Kroah-Hartman
2016-01-20 22:00 ` [PATCH 3.14 21/47] ext4, jbd2: ensure entering into panic after recording an error in superblock Greg Kroah-Hartman
2016-01-20 22:00 ` [PATCH 3.14 22/47] firewire: ohci: fix JMicron JMB38x IT context discovery Greg Kroah-Hartman
2016-01-20 22:00 ` [PATCH 3.14 23/47] nfs4: start callback_ident at idr 1 Greg Kroah-Hartman
2016-01-20 22:00 ` [PATCH 3.14 24/47] nfs: if we have no valid attrs, then dont declare the attribute cache valid Greg Kroah-Hartman
2016-01-20 22:00 ` [PATCH 3.14 25/47] ocfs2: fix umask ignored issue Greg Kroah-Hartman
2016-01-20 22:00 ` [PATCH 3.14 26/47] USB: cdc_acm: Ignore Infineon Flash Loader utility Greg Kroah-Hartman
2016-01-20 22:00 ` [PATCH 3.14 27/47] USB: serial: Another Infineon flash loader USB ID Greg Kroah-Hartman
2016-01-20 22:01 ` [PATCH 3.14 28/47] USB: cp210x: Remove CP2110 ID from compatibility list Greg Kroah-Hartman
2016-01-20 22:01 ` [PATCH 3.14 29/47] USB: add quirk for devices with broken LPM Greg Kroah-Hartman
2016-01-20 22:01 ` [PATCH 3.14 30/47] USB: whci-hcd: add check for dma mapping error Greg Kroah-Hartman
2016-01-20 22:01 ` [PATCH 3.14 31/47] usb: Use the USB_SS_MULT() macro to decode burst multiplier for log message Greg Kroah-Hartman
2016-01-20 22:01 ` [PATCH 3.14 32/47] gre6: allow to update all parameters via rtnl Greg Kroah-Hartman
2016-01-20 22:01 ` [PATCH 3.14 33/47] atl1c: Improve driver not to do order 4 GFP_ATOMIC allocation Greg Kroah-Hartman
2016-01-20 22:01 ` [PATCH 3.14 34/47] sctp: use the same clock as if sock source timestamps were on Greg Kroah-Hartman
2016-01-20 22:01 ` [PATCH 3.14 35/47] sctp: update the netstamp_needed counter when copying sockets Greg Kroah-Hartman
2016-01-20 22:01 ` [PATCH 3.14 36/47] ipv6: sctp: clone options to avoid use after free Greg Kroah-Hartman
2016-01-20 22:01 ` [PATCH 3.14 38/47] sh_eth: fix kernel oops in skb_put() Greg Kroah-Hartman
2016-01-20 22:01 ` [PATCH 3.14 39/47] vlan: Fix untag operations of stacked vlans with REORDER_HEADER off Greg Kroah-Hartman
2016-01-20 22:01 ` [PATCH 3.14 40/47] skbuff: Fix offset error in skb_reorder_vlan_header Greg Kroah-Hartman
2016-01-20 22:01 ` [PATCH 3.14 41/47] pptp: verify sockaddr_len in pptp_bind() and pptp_connect() Greg Kroah-Hartman
2016-01-20 22:01 ` [PATCH 3.14 42/47] bluetooth: Validate socket address length in sco_sock_bind() Greg Kroah-Hartman
2016-01-20 22:01 ` [PATCH 3.14 43/47] af_unix: Revert lock_interruptible in stream receive code Greg Kroah-Hartman
2016-01-20 22:01 ` [PATCH 3.14 44/47] KEYS: Fix race between key destruction and finding a keyring by name Greg Kroah-Hartman
2016-01-20 22:01 ` [PATCH 3.14 45/47] KEYS: Fix crash when attempt to garbage collect an uninstantiated keyring Greg Kroah-Hartman
2016-01-20 22:01 ` [PATCH 3.14 46/47] KEYS: Fix race between read and revoke Greg Kroah-Hartman
2016-01-20 22:01 ` [PATCH 3.14 47/47] KEYS: Fix keyring ref leak in join_session_keyring() Greg Kroah-Hartman
2016-01-20 23:15 ` [PATCH 3.14 00/47] 3.14.59-stable review Shuah Khan
2016-01-21 12:21 ` Guenter Roeck

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=20160120215508.714083342@linuxfoundation.org \
    --to=gregkh@linuxfoundation.org \
    --cc=Jason@zx2c4.com \
    --cc=davem@davemloft.net \
    --cc=hannes@stressinduktion.org \
    --cc=linux-kernel@vger.kernel.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;
as well as URLs for NNTP newsgroup(s).