mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Ben Hutchings <ben@decadent.org.uk>
To: linux-kernel@vger.kernel.org, stable@vger.kernel.org
Cc: akpm@linux-foundation.org,
	"Paul E. McKenney" <paulmck@linux.vnet.ibm.com>,
	"Eric Dumazet" <eric.dumazet@gmail.com>,
	"Eric Dumazet" <edumazet@google.com>,
	"Boris Zhmurov" <zhmurov@yandex-team.ru>,
	"Roman Gushchin" <klamm@yandex-team.ru>,
	"David S. Miller" <davem@davemloft.net>
Subject: [20/26] net: force a reload of first item in  hlist_nulls_for_each_entry_rcu
Date: Wed, 26 Jun 2013 03:35:59 +0100	[thread overview]
Message-ID: <lsq.1372214159.708140691@decadent.org.uk> (raw)
In-Reply-To: <lsq.1372214158.425780530@decadent.org.uk>

3.2.48-rc1 review patch.  If anyone has any objections, please let me know.

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

From: Eric Dumazet <eric.dumazet@gmail.com>

[ Upstream commit c87a124a5d5e8cf8e21c4363c3372bcaf53ea190 ]

Roman Gushchin discovered that udp4_lib_lookup2() was not reloading
first item in the rcu protected list, in case the loop was restarted.

This produced soft lockups as in https://lkml.org/lkml/2013/4/16/37

rcu_dereference(X)/ACCESS_ONCE(X) seem to not work as intended if X is
ptr->field :

In some cases, gcc caches the value or ptr->field in a register.

Use a barrier() to disallow such caching, as documented in
Documentation/atomic_ops.txt line 114

Thanks a lot to Roman for providing analysis and numerous patches.

Diagnosed-by: Roman Gushchin <klamm@yandex-team.ru>
Signed-off-by: Eric Dumazet <edumazet@google.com>
Reported-by: Boris Zhmurov <zhmurov@yandex-team.ru>
Signed-off-by: Roman Gushchin <klamm@yandex-team.ru>
Acked-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Ben Hutchings <ben@decadent.org.uk>
---
 include/linux/rculist_nulls.h | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)

diff --git a/include/linux/rculist_nulls.h b/include/linux/rculist_nulls.h
index 2ae1371..1c33dd7 100644
--- a/include/linux/rculist_nulls.h
+++ b/include/linux/rculist_nulls.h
@@ -105,9 +105,14 @@ static inline void hlist_nulls_add_head_rcu(struct hlist_nulls_node *n,
  * @head:	the head for your list.
  * @member:	the name of the hlist_nulls_node within the struct.
  *
+ * The barrier() is needed to make sure compiler doesn't cache first element [1],
+ * as this loop can be restarted [2]
+ * [1] Documentation/atomic_ops.txt around line 114
+ * [2] Documentation/RCU/rculist_nulls.txt around line 146
  */
 #define hlist_nulls_for_each_entry_rcu(tpos, pos, head, member)			\
-	for (pos = rcu_dereference_raw(hlist_nulls_first_rcu(head));		\
+	for (({barrier();}),							\
+	     pos = rcu_dereference_raw(hlist_nulls_first_rcu(head));		\
 		(!is_a_nulls(pos)) &&						\
 		({ tpos = hlist_nulls_entry(pos, typeof(*tpos), member); 1; }); \
 		pos = rcu_dereference_raw(hlist_nulls_next_rcu(pos)))


  parent reply	other threads:[~2013-06-26  2:47 UTC|newest]

Thread overview: 35+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-06-26  2:35 [00/26] 3.2.48-rc1 review Ben Hutchings
2013-06-26  2:35 ` [19/26] net: Block MSG_CMSG_COMPAT in send(m)msg and recv(m)msg Ben Hutchings
2013-06-26  2:35 ` [11/26] virtio-blk: Call revalidate_disk() upon online disk resize Ben Hutchings
2013-06-26  2:35 ` [12/26] tcp: fix tcp_md5_hash_skb_data() Ben Hutchings
2013-06-26  2:35 ` [09/26] x86/efi: Fix dummy variable buffer allocation Ben Hutchings
2013-06-26  2:35 ` [17/26] tcp: xps: fix reordering issues Ben Hutchings
2013-06-26  2:35 ` [02/26] ALSA: usb-audio: work around Android accessory firmware bug Ben Hutchings
2013-06-26  2:35 ` [01/26] tilepro: work around module link error with gcc 4.7 Ben Hutchings
2013-06-26  2:35 ` [04/26] ARM: 7755/1: handle user space mapped pages in flush_kernel_dcache_page Ben Hutchings
2013-06-26  9:04   ` Luis Henriques
2013-06-26 22:35     ` Simon Baatz
2013-06-27  8:21       ` Luis Henriques
2013-06-29  3:01       ` Ben Hutchings
2013-06-26  2:35 ` [13/26] gianfar: add missing iounmap() on error in gianfar_ptp_probe() Ben Hutchings
2013-06-26  2:35 ` [14/26] ipv6: fix possible crashes in ip6_cork_release() Ben Hutchings
2013-06-26  2:35 ` [06/26] x86: fix build error and kconfig for ia32_emulation and binfmt Ben Hutchings
2013-06-26  2:35 ` [18/26] ip_tunnel: fix kernel panic with icmp_dest_unreach Ben Hutchings
2013-06-26  2:35 ` [05/26] KVM: x86: remove vcpu's CPL check in host-invoked XCR set Ben Hutchings
2013-06-26  2:35 ` [03/26] ALSA: usb-audio: Fix invalid volume resolution for Logitech HD Webcam c310 Ben Hutchings
2013-06-26  2:35 ` [15/26] netlabel: improve domain mapping validation Ben Hutchings
2013-06-26  2:35 ` [10/26] Revert "drm/i915: GFX_MODE Flush TLB Invalidate Mode must be '1' for scanline waits" Ben Hutchings
2013-06-26  2:35 ` [07/26] USB: serial: ti_usb_3410_5052: new device id for Abbot strip port cable Ben Hutchings
2013-06-26  8:21   ` Anders Hammarquist
2013-06-26 17:22     ` Greg Kroah-Hartman
2013-06-26  2:35 ` [16/26] r8169: fix 8168evl frame padding Ben Hutchings
2013-06-26  8:51   ` Stefan Bader
2013-06-26 18:37     ` David Miller
2013-06-26 20:00       ` Stefan Bader
2013-06-26  2:35 ` [21/26] ipv6: assign rt6_info to inet6_ifaddr in init_loopback Ben Hutchings
2013-06-26  2:35 ` [24/26] l2tp: Fix PPP header erasure and memory leak Ben Hutchings
2013-06-26  2:35 ` [22/26] net: sctp: fix NULL pointer dereference in socket destruction Ben Hutchings
2013-06-26  2:35 ` [23/26] packet: packet_getname_spkt: make sure string is always 0-terminated Ben Hutchings
2013-06-26  2:35 ` [26/26] ncpfs: fix rmdir returns Device or resource busy Ben Hutchings
2013-06-26  2:35 ` Ben Hutchings [this message]
2013-06-26  2:35 ` [25/26] l2tp: Fix sendmsg() return value Ben Hutchings

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=lsq.1372214159.708140691@decadent.org.uk \
    --to=ben@decadent.org.uk \
    --cc=akpm@linux-foundation.org \
    --cc=davem@davemloft.net \
    --cc=edumazet@google.com \
    --cc=eric.dumazet@gmail.com \
    --cc=klamm@yandex-team.ru \
    --cc=linux-kernel@vger.kernel.org \
    --cc=paulmck@linux.vnet.ibm.com \
    --cc=stable@vger.kernel.org \
    --cc=zhmurov@yandex-team.ru \
    /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

all inboxes | Powered by JetHome®