From: Darren Hart <dvhart@linux.intel.com>
To: Siddhesh Poyarekar <siddhesh.poyarekar@gmail.com>
Cc: linux-kernel@vger.kernel.org, Thomas Gleixner <tglx@linutronix.de>
Subject: Re: [PATCH] [RESEND 2] Take over futex of dead task only if FUTEX_WAITERS is not set
Date: Tue, 23 Oct 2012 07:04:01 -0700 [thread overview]
Message-ID: <5086A3D1.7080709@linux.intel.com> (raw)
In-Reply-To: <1350876034-22023-1-git-send-email-siddhesh.poyarekar@gmail.com>
Hi Siddesh,
Thanks for the patch and your work to isolate it in the glibc bug 14076.
On 10/21/2012 08:20 PM, Siddhesh Poyarekar wrote:
> In futex_lock_pi_atomic, we consider that if the value in the futex
> variable is 0 with additional flags, then it is safe for takeover
> since the owner of the futex is dead. However, when FUTEX_WAITERS is
> set in the futex value, handle_futex_death calls futex_wake to wake up
> one task.
It shouldn't for PI mutexes. It should just set the FUTEX_OWNER_DIED flag,
maintaining the FUTEX_WAITERS flag, and exit.
int handle_futex_death(...
...
/*
* Wake robust non-PI futexes here. The wakeup of
* PI futexes happens in exit_pi_state():
*/
if (!pi && (uval & FUTEX_WAITERS))
futex_wake(uaddr, 1, 1, FUTEX_BITSET_MATCH_ANY);
There may still be an issue, as the commentary around exit_pi_state_list()
doesn't inspire confidence.
--
Darren
> Hence the assumption in futex_lock_pi_atomic is not correct.
> The correct assumption is that a futex may be considered safe for a
> takeover if The FUTEX_OWNER_DIED bit is set, the TID bits are 0 and
> the FUTEX_WAITERS bit is not set.
>
> The race described above can be seen in the reproducer in the
> following glibc bug report:
>
> http://sourceware.org/bugzilla/show_bug.cgi?id=14076
>
> Signed-off-by: Siddhesh Poyarekar <siddhesh.poyarekar@gmail.com>
> ---
> kernel/futex.c | 7 ++++++-
> 1 files changed, 6 insertions(+), 1 deletions(-)
>
> diff --git a/kernel/futex.c b/kernel/futex.c
> index 3717e7b..9aa2d5a 100644
> --- a/kernel/futex.c
> +++ b/kernel/futex.c
> @@ -760,9 +760,14 @@ retry:
> * case. We also do an unconditional take over, when the owner
> * of the futex died.
> *
> + * We do not take over the futex if FUTEX_WAITERS is set because we
> + * could end up waking two tasks, the current one and the one that the
> + * futex death event wakes in handle_futex_death.
> + *
> * This is safe as we are protected by the hash bucket lock !
> */
> - if (unlikely(ownerdied || !(curval & FUTEX_TID_MASK))) {
> + if (unlikely(ownerdied ||
> + !(curval & (FUTEX_TID_MASK | FUTEX_WAITERS)))) {
> /* Keep the OWNER_DIED bit */
> newval = (curval & ~FUTEX_TID_MASK) | vpid;
> ownerdied = 0;
>
--
Darren Hart
Intel Open Source Technology Center
Yocto Project - Technical Lead - Linux Kernel
next prev parent reply other threads:[~2012-10-23 14:35 UTC|newest]
Thread overview: 13+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-10-11 14:52 [PATCH] Take over futex of dead task only if FUTEX_WAITERS is not set Siddhesh Poyarekar
2012-10-17 7:15 ` [PATCH RESEND] " Siddhesh Poyarekar
2012-10-22 3:20 ` [PATCH] [RESEND 2] " Siddhesh Poyarekar
2012-10-23 14:04 ` Darren Hart [this message]
2012-10-23 20:29 ` Thomas Gleixner
2012-10-24 12:48 ` Siddhesh Poyarekar
2012-10-24 18:08 ` Thomas Gleixner
2012-10-25 4:36 ` Darren Hart
2012-10-25 4:44 ` Siddhesh Poyarekar
2012-10-25 4:33 ` Darren Hart
2012-10-25 8:14 ` Thomas Gleixner
2012-10-25 8:18 ` Darren Hart
2012-11-01 21:35 ` [tip:core/urgent] futex: Handle futex_pi OWNER_DIED take over correctly tip-bot for Thomas Gleixner
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=5086A3D1.7080709@linux.intel.com \
--to=dvhart@linux.intel.com \
--cc=linux-kernel@vger.kernel.org \
--cc=siddhesh.poyarekar@gmail.com \
--cc=tglx@linutronix.de \
/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).