linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To: linux-kernel@vger.kernel.org, stable@vger.kernel.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	alan@lxorguk.ukuu.org.uk, Darren Hart <dvhart@linux.intel.com>,
	Dave Jones <davej@redat.com>,
	Thomas Gleixner <tglx@linutronix.de>,
	Peter Zijlstra <peterz@infradead.org>,
	Ingo Molnar <mingo@elte.hu>, John Kacur <jkacur@redhat.com>,
	Andrew Morton <akpm@linux-foundation.org>,
	Linus Torvalds <torvalds@linux-foundation.org>
Subject: [ 17/37] futex: avoid wake_futex() for a PI futex_q
Date: Fri, 30 Nov 2012 10:46:04 -0800	[thread overview]
Message-ID: <20121130183859.088336238@linuxfoundation.org> (raw)
In-Reply-To: <20121130183857.166228045@linuxfoundation.org>

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

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

From: Darren Hart <dvhart@linux.intel.com>

commit aa10990e028cac3d5e255711fb9fb47e00700e35 upstream.

Dave Jones reported a bug with futex_lock_pi() that his trinity test
exposed.  Sometime between queue_me() and taking the q.lock_ptr, the
lock_ptr became NULL, resulting in a crash.

While futex_wake() is careful to not call wake_futex() on futex_q's with
a pi_state or an rt_waiter (which are either waiting for a
futex_unlock_pi() or a PI futex_requeue()), futex_wake_op() and
futex_requeue() do not perform the same test.

Update futex_wake_op() and futex_requeue() to test for q.pi_state and
q.rt_waiter and abort with -EINVAL if detected.  To ensure any future
breakage is caught, add a WARN() to wake_futex() if the same condition
is true.

This fix has seen 3 hours of testing with "trinity -c futex" on an
x86_64 VM with 4 CPUS.

[akpm@linux-foundation.org: tidy up the WARN()]
Signed-off-by: Darren Hart <dvhart@linux.intel.com>
Reported-by: Dave Jones <davej@redat.com>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: John Kacur <jkacur@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

---
 kernel/futex.c |   18 +++++++++++++++++-
 1 file changed, 17 insertions(+), 1 deletion(-)

--- a/kernel/futex.c
+++ b/kernel/futex.c
@@ -843,6 +843,9 @@ static void wake_futex(struct futex_q *q
 {
 	struct task_struct *p = q->task;
 
+	if (WARN(q->pi_state || q->rt_waiter, "refusing to wake PI futex\n"))
+		return;
+
 	/*
 	 * We set q->lock_ptr = NULL _before_ we wake up the task. If
 	 * a non-futex wake up happens on another CPU then the task
@@ -1078,6 +1081,10 @@ retry_private:
 
 	plist_for_each_entry_safe(this, next, head, list) {
 		if (match_futex (&this->key, &key1)) {
+			if (this->pi_state || this->rt_waiter) {
+				ret = -EINVAL;
+				goto out_unlock;
+			}
 			wake_futex(this);
 			if (++ret >= nr_wake)
 				break;
@@ -1090,6 +1097,10 @@ retry_private:
 		op_ret = 0;
 		plist_for_each_entry_safe(this, next, head, list) {
 			if (match_futex (&this->key, &key2)) {
+				if (this->pi_state || this->rt_waiter) {
+					ret = -EINVAL;
+					goto out_unlock;
+				}
 				wake_futex(this);
 				if (++op_ret >= nr_wake2)
 					break;
@@ -1098,6 +1109,7 @@ retry_private:
 		ret += op_ret;
 	}
 
+out_unlock:
 	double_unlock_hb(hb1, hb2);
 out_put_keys:
 	put_futex_key(&key2);
@@ -1387,9 +1399,13 @@ retry_private:
 		/*
 		 * FUTEX_WAIT_REQEUE_PI and FUTEX_CMP_REQUEUE_PI should always
 		 * be paired with each other and no other futex ops.
+		 *
+		 * We should never be requeueing a futex_q with a pi_state,
+		 * which is awaiting a futex_unlock_pi().
 		 */
 		if ((requeue_pi && !this->rt_waiter) ||
-		    (!requeue_pi && this->rt_waiter)) {
+		    (!requeue_pi && this->rt_waiter) ||
+		    this->pi_state) {
 			ret = -EINVAL;
 			break;
 		}



  parent reply	other threads:[~2012-11-30 19:33 UTC|newest]

Thread overview: 50+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-11-30 18:45 [ 00/37] 3.0.54-stable review Greg Kroah-Hartman
2012-11-30 18:45 ` [ 01/37] ALSA: pcmcia - Use pcmcia_request_irq() Greg Kroah-Hartman
2012-11-30 18:45 ` [ 02/37] drivers/block/DAC960: fix DAC960_V2_IOCTL_Opcode_T -Wenum-compare warning Greg Kroah-Hartman
2012-11-30 18:45 ` [ 03/37] drivers/block/DAC960: fix -Wuninitialized warning Greg Kroah-Hartman
2012-11-30 18:45 ` [ 04/37] riva/fbdev: fix several -Wuninitialized Greg Kroah-Hartman
2012-11-30 18:45 ` [ 05/37] ifenslave: Fix unused variable warnings Greg Kroah-Hartman
2012-11-30 18:45 ` [ 06/37] x86-32: Fix invalid stack address while in softirq Greg Kroah-Hartman
2012-12-04 13:42   ` Herton Ronaldo Krzesinski
2012-12-04 14:13     ` Robert Richter
2012-12-06 18:42     ` Greg Kroah-Hartman
2012-11-30 18:45 ` [ 07/37] x86, microcode, AMD: Add support for family 16h processors Greg Kroah-Hartman
2012-11-30 18:45 ` [ 08/37] rtlwifi: rtl8192cu: Add new USB ID Greg Kroah-Hartman
2012-11-30 18:45 ` [ 09/37] mwifiex: report error to MMC core if we cannot suspend Greg Kroah-Hartman
2012-11-30 18:45 ` [ 10/37] SCSI: isci: copy fis 0x34 response into proper buffer Greg Kroah-Hartman
2012-11-30 18:45 ` [ 11/37] ALSA: ua101, usx2y: fix broken MIDI output Greg Kroah-Hartman
2012-11-30 18:45 ` [ 12/37] ALSA: hda - Cirrus: Correctly clear line_out_pins when moving to speaker Greg Kroah-Hartman
2012-12-03  9:46   ` David Henningsson
2012-12-03 20:56     ` Greg Kroah-Hartman
2012-11-30 18:46 ` [ 13/37] PARISC: fix virtual aliasing issue in get_shared_area() Greg Kroah-Hartman
2012-11-30 18:46 ` [ 14/37] PARISC: fix user-triggerable panic on parisc Greg Kroah-Hartman
2012-11-30 18:46 ` [ 15/37] mtd: slram: invalid checking of absolute end address Greg Kroah-Hartman
2012-11-30 18:46 ` [ 16/37] dm: fix deadlock with request based dm and queue request_fn recursion Greg Kroah-Hartman
2012-11-30 18:46 ` Greg Kroah-Hartman [this message]
2012-11-30 18:46 ` [ 18/37] mac80211: deinitialize ibss-internals after emptiness check Greg Kroah-Hartman
2012-11-30 18:46 ` [ 19/37] radeon: add AGPMode 1 quirk for RV250 Greg Kroah-Hartman
2012-11-30 18:46 ` [ 20/37] can: bcm: initialize ifindex for timeouts without previous frame reception Greg Kroah-Hartman
2012-11-30 18:46 ` [ 21/37] jbd: Fix lock ordering bug in journal_unmap_buffer() Greg Kroah-Hartman
2012-11-30 18:46 ` [ 22/37] sparc64: not any error from do_sigaltstack() should fail rt_sigreturn() Greg Kroah-Hartman
2012-11-30 18:46 ` [ 23/37] ALSA: hda - Add new codec ALC283 ALC290 support Greg Kroah-Hartman
2012-11-30 18:46 ` [ 24/37] ALSA: hda - Fix missing beep on ASUS X43U notebook Greg Kroah-Hartman
2012-11-30 18:46 ` [ 25/37] ALSA: hda - Add support for Realtek ALC292 Greg Kroah-Hartman
2012-11-30 18:46 ` [ 26/37] bas_gigaset: fix pre_reset handling Greg Kroah-Hartman
2012-11-30 18:46 ` [ 27/37] ixgbe: add support for X540-AT1 Greg Kroah-Hartman
2012-11-30 18:46 ` [ 28/37] sata_svw: check DMA start bit before reset Greg Kroah-Hartman
2012-11-30 18:46 ` [ 29/37] ixgbe: add support for new 82599 device Greg Kroah-Hartman
2012-11-30 18:46 ` [ 30/37] ixgbe: add support for new 82599 device id Greg Kroah-Hartman
2012-11-30 18:46 ` [ 31/37] get_dvb_firmware: fix download site for tda10046 firmware Greg Kroah-Hartman
2012-11-30 18:46 ` [ 32/37] USB: mct_u232: fix broken close Greg Kroah-Hartman
2012-11-30 18:46 ` [ 33/37] watchdog: using u64 in get_sample_period() Greg Kroah-Hartman
2012-11-30 18:46 ` [ 34/37] acer-wmi: support for P key on TM8372 Greg Kroah-Hartman
2012-11-30 18:46 ` [ 35/37] x86, mce, therm_throt: Dont report power limit and package level thermal throttle events in mcelog Greg Kroah-Hartman
2012-11-30 18:46 ` [ 36/37] Input: bcm5974 - set BUTTONPAD property Greg Kroah-Hartman
2012-11-30 18:46 ` [ 37/37] mmc: sdhci-s3c: fix the wrong number of max bus clocks Greg Kroah-Hartman
2012-12-01 15:36 ` [ 00/37] 3.0.54-stable review Satoru Takeuchi
2012-12-01 16:24   ` David Miller
2012-12-01 17:15     ` Shuah Khan
2012-12-02  0:27       ` Satoru Takeuchi
2012-12-02  2:07 ` Shuah Khan
2012-12-02 17:01   ` Greg Kroah-Hartman
2012-12-02 19:36 ` Nikola Ciprich

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=20121130183859.088336238@linuxfoundation.org \
    --to=gregkh@linuxfoundation.org \
    --cc=akpm@linux-foundation.org \
    --cc=alan@lxorguk.ukuu.org.uk \
    --cc=davej@redat.com \
    --cc=dvhart@linux.intel.com \
    --cc=jkacur@redhat.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@elte.hu \
    --cc=peterz@infradead.org \
    --cc=stable@vger.kernel.org \
    --cc=tglx@linutronix.de \
    --cc=torvalds@linux-foundation.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).