public inbox for stable@vger.kernel.org
 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, "Kees Cook" <keescook@chromium.org>,
	"Linus Torvalds" <torvalds@linux-foundation.org>,
	"Darren Hart" <dvhart@linux.intel.com>,
	"Thomas Gleixner" <tglx@linutronix.de>,
	"Will Drewry" <wad@chromium.org>,
	Pinkie@decadent.org.uk
Subject: [PATCH 3.2 89/92] futex-prevent-requeue-pi-on-same-futex.patch futex: Forbid uaddr == uaddr2 in futex_requeue(..., requeue_pi=1)
Date: Sat, 07 Jun 2014 02:26:29 +0100	[thread overview]
Message-ID: <lsq.1402104389.125698523@decadent.org.uk> (raw)
In-Reply-To: <lsq.1402104388.385007124@decadent.org.uk>

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

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

From: Thomas Gleixner <tglx@linutronix.de>

commit e9c243a5a6de0be8e584c604d353412584b592f8 upstream.

If uaddr == uaddr2, then we have broken the rule of only requeueing from
a non-pi futex to a pi futex with this call.  If we attempt this, then
dangling pointers may be left for rt_waiter resulting in an exploitable
condition.

This change brings futex_requeue() in line with futex_wait_requeue_pi()
which performs the same check as per commit 6f7b0a2a5c0f ("futex: Forbid
uaddr == uaddr2 in futex_wait_requeue_pi()")

[ tglx: Compare the resulting keys as well, as uaddrs might be
  	different depending on the mapping ]

Fixes CVE-2014-3153.

Reported-by: Pinkie Pie
Signed-off-by: Will Drewry <wad@chromium.org>
Signed-off-by: Kees Cook <keescook@chromium.org>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Darren Hart <dvhart@linux.intel.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Ben Hutchings <ben@decadent.org.uk>
---
 kernel/futex.c | 25 +++++++++++++++++++++++++
 1 file changed, 25 insertions(+)

--- a/kernel/futex.c
+++ b/kernel/futex.c
@@ -1289,6 +1289,13 @@ static int futex_requeue(u32 __user *uad
 
 	if (requeue_pi) {
 		/*
+		 * Requeue PI only works on two distinct uaddrs. This
+		 * check is only valid for private futexes. See below.
+		 */
+		if (uaddr1 == uaddr2)
+			return -EINVAL;
+
+		/*
 		 * requeue_pi requires a pi_state, try to allocate it now
 		 * without any locks in case it fails.
 		 */
@@ -1326,6 +1333,15 @@ retry:
 	if (unlikely(ret != 0))
 		goto out_put_key1;
 
+	/*
+	 * The check above which compares uaddrs is not sufficient for
+	 * shared futexes. We need to compare the keys:
+	 */
+	if (requeue_pi && match_futex(&key1, &key2)) {
+		ret = -EINVAL;
+		goto out_put_keys;
+	}
+
 	hb1 = hash_futex(&key1);
 	hb2 = hash_futex(&key2);
 
@@ -2357,6 +2373,15 @@ static int futex_wait_requeue_pi(u32 __u
 	if (ret)
 		goto out_key2;
 
+	/*
+	 * The check above which compares uaddrs is not sufficient for
+	 * shared futexes. We need to compare the keys:
+	 */
+	if (match_futex(&q.key, &key2)) {
+		ret = -EINVAL;
+		goto out_put_keys;
+	}
+
 	/* Queue the futex_q, drop the hb lock, wait for wakeup. */
 	futex_wait_queue_me(hb, &q, to);
 


  parent reply	other threads:[~2014-06-07  1:26 UTC|newest]

Thread overview: 98+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-06-07  1:26 [PATCH 3.2 00/92] 3.2.60-rc1 review Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 19/92] tcp_cubic: fix the range of delayed_ack Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 17/92] sctp: reset flowi4_oif parameter on route lookup Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 30/92] drm/vmwgfx: Make sure user-space can't DMA across buffer object boundaries v2 Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 20/92] net: ipv4: ip_forward: fix inverted local_df test Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 07/92] l2tp: take PMTU from tunnel UDP socket Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 24/92] ipv4: initialise the itag variable in __mkroute_input Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 06/92] tracepoint: Do not waste memory on mods with no tracepoints Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 21/92] ipv4: fib_semantics: increment fib_info_cnt after fib_info allocation Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 09/92] bonding: Remove debug_fs files when module init fails Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 18/92] Revert "macvlan : fix checksums error when we are in bridge mode" Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 01/92] powerpc: Add vr save/restore functions Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 25/92] skb: Add inline helper for getting the skb end offset from head Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 11/92] net: ipv4: current group_info should be put after using Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 03/92] ACPI / EC: Process rather than discard events in acpi_ec_clear Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 05/92] netfilter: Can't fail and free after table replacement Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 23/92] act_mirred: do not drop packets when fails to mirror it Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 29/92] Input: synaptics - add min/max quirk for ThinkPad Edge E431 Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 08/92] net: core: don't account for udp header size when computing seglen Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 04/92] tgafb: fix mode setting with fbset Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 14/92] rtnetlink: Warn when interface's information won't fit in our packet Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 26/92] net-gro: reset skb->truesize in napi_reuse_skb() Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 28/92] rt2x00: fix beaconing on USB Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 10/92] ipv6: Limit mtu to 65575 bytes Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 16/92] bridge: Handle IFLA_ADDRESS correctly when creating bridge device Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 02/92] ACPI / EC: Clear stale EC events on Samsung systems Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 13/92] tg3: update rx_jumbo_pending ring param only when jumbo frames are enabled Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 15/92] rtnetlink: Only supply IFLA_VF_PORTS information when RTEXT_FILTER_VF is set Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 27/92] thinkpad-acpi: fix issuing duplicated key events for brightness up/down Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 22/92] macvlan: Don't propagate IFF_ALLMULTI changes on down interfaces Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 12/92] filter: prevent nla extensions to peek beyond the end of the message Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 63/92] NFSD: Call ->set_acl with a NULL ACL structure if no entries Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 43/92] rtl8192cu: Fix unbalanced irq enable in error path of rtl92cu_hw_init() Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 39/92] hrtimer: Prevent all reprogramming if hang detected Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 36/92] KVM: async_pf: mm->mm_users can not pin apf->mm Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 88/92] futex: Prevent attaching to kernel threads Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 54/92] Input: elantech - fix touchpad initialization on Gigabyte U2442 Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 80/92] can: peak_pci: Fix the way channels are linked together Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 59/92] Negative (setpoint-dirty) in bdi_position_ratio() Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 77/92] perf: Prevent false warning in perf_swevent_add Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 35/92] kvm: remove .done from struct kvm_async_pf Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 73/92] PCI: shpchp: Check bridge's secondary (not primary) bus speed Ben Hutchings
2014-06-07  1:26 ` Ben Hutchings [this message]
2014-06-07  1:26 ` [PATCH 3.2 53/92] HID: add NO_INIT_REPORTS quirk for Synaptics Touch Pad V 103S Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 67/92] [media] V4L2: ov7670: fix a wrong index, potentially Oopsing the kernel from user-space Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 78/92] drm/radeon: also try GART for CPU accessed buffers Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 51/92] HID: usbhid: quirk for Synaptics HD touchscreen Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 55/92] md: avoid possible spinning md thread at shutdown Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 86/92] mm/memory-failure.c: fix memory leak by race between poison and unpoison Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 61/92] sched: Use CPUPRI_NR_PRIORITIES instead of MAX_RT_PRIO in cpupri check Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 76/92] perf: Limit perf_event_attr::sample_period to 63 bits Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 74/92] libceph: only call kernel_sendpage() via helper Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 82/92] nfsd4: remove lockowner when removing lock stateid Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 68/92] [media] V4L2: fix VIDIOC_CREATE_BUFS in 64- / 32-bit compatibility mode Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 60/92] mm/page-writeback.c: fix divide by zero in pos_ratio_polynom Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 71/92] i2c: s3c2410: resume race fix Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 48/92] USB: Nokia 305 should be treated as unusual dev Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 46/92] usb: storage: shuttle_usbat: fix discs being detected twice Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 52/92] HID: usbhid: quirk for Synaptics Quad HD touchscreen Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 90/92] futex: Validate atomic acquisition in futex_lock_pi_atomic() Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 75/92] libceph: fix corruption when using page_count 0 page in rbd Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 38/92] [SCSI] mpt2sas: Don't disable device twice at suspend Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 64/92] hrtimer: Set expiry time before switch_hrtimer_base() Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 72/92] x86-64, modify_ldt: Make support for 16-bit segments a runtime option Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 31/92] Bluetooth: Fix redundant encryption request for reauthentication Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 83/92] nfsd4: warn on finding lockowner without stateid's Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 87/92] futex: Add another early deadlock detection check Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 32/92] Bluetooth: Add support for Lite-on [04ca:3007] Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 50/92] HID: usbhid: quirk for Synaptics Large Touchccreen Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 62/92] trace: module: Maintain a valid user count Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 47/92] drivers/tty/hvc: don't free hvc_console_setup after init Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 69/92] x86, mm, hugetlb: Add missing TLB page invalidation for hugetlb_cow() Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 58/92] posix_acl: handle NULL ACL in posix_acl_equiv_mode Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 92/92] futex: Make lookup_pi_state more robust Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 57/92] NFSd: call rpc_destroy_wait_queue() from free_client() Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 49/92] USB: Nokia 5300 should be treated as unusual dev Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 91/92] futex: Always cleanup owner tid in unlock_pi Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 79/92] drm/radeon: handle non-VGA class pci devices with ATRM Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 84/92] dma: mv_xor: Flush descriptors before activating a channel Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 85/92] hwpoison, hugetlb: lock_page/unlock_page does not match for handling a free hugepage Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 42/92] rtlwifi: rtl8192cu: Fix too long disable of IRQs Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 33/92] crypto: caam - add allocation failure handling in SPRINTFCAT macro Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 44/92] [media] media-device: fix infoleak in ioctl media_enum_entities() Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 81/92] can: peak_pci: prevent use after free at netdev removal Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 41/92] timer: Prevent overflow in apply_slack Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 70/92] i2c: designware: Mask all interrupts during i2c controller enable Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 66/92] hwmon: (emc1403) Support full range of known chip revision numbers Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 40/92] hrtimer: Prevent remote enqueue of leftmost timers Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 45/92] drm/nouveau/acpi: allow non-optimus setups to load vbios from acpi Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 34/92] kvm: free resources after canceling async_pf Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 56/92] NFSd: Move default initialisers from create_client() to alloc_client() Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 65/92] hwmon: (emc1403) fix inverted store_hyst() Ben Hutchings
2014-06-07  1:26 ` [PATCH 3.2 37/92] ftrace/module: Hardcode ftrace_module_init() call into load_module() Ben Hutchings
2014-06-07  2:23 ` [PATCH 3.2 00/92] 3.2.60-rc1 review Ben Hutchings
2014-06-07 16:33 ` Guenter Roeck
2014-06-07 17:00   ` Ben Hutchings
2014-06-09 23:25     ` Satoru Takeuchi
2014-06-09 23:48       ` 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.1402104389.125698523@decadent.org.uk \
    --to=ben@decadent.org.uk \
    --cc=Pinkie@decadent.org.uk \
    --cc=akpm@linux-foundation.org \
    --cc=dvhart@linux.intel.com \
    --cc=keescook@chromium.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=stable@vger.kernel.org \
    --cc=tglx@linutronix.de \
    --cc=torvalds@linux-foundation.org \
    --cc=wad@chromium.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