[PATCH 4.4.257 0/1] Bugfix for ad4740ceccfb ("futex: Avoid violating the 10th rule of futex")

[PATCH 4.4.257 0/1] Bugfix for ad4740ceccfb ("futex: Avoid violating the 10th rule of futex")

[Zheng Yejian]

*** BLURB HERE ***

Peter Zijlstra (1):
  futex: Fix OWNER_DEAD fixup

 kernel/futex.c | 7 +++----
 1 file changed, 3 insertions(+), 4 deletions(-)

-- 
2.25.4

[PATCH 1/1] futex: Fix OWNER_DEAD fixup

[Zheng Yejian]

From: Peter Zijlstra <peterz@infradead.org>

commit a97cb0e7b3f4c6297fd857055ae8e895f402f501 upstream.

Both Geert and DaveJ reported that the recent futex commit:

  c1e2f0eaf015 ("futex: Avoid violating the 10th rule of futex")

introduced a problem with setting OWNER_DEAD. We set the bit on an
uninitialized variable and then entirely optimize it away as a
dead-store.

Move the setting of the bit to where it is more useful.

Reported-by: Geert Uytterhoeven <geert@linux-m68k.org>
Reported-by: Dave Jones <davej@codemonkey.org.uk>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Paul E. McKenney <paulmck@us.ibm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Fixes: c1e2f0eaf015 ("futex: Avoid violating the 10th rule of futex")
Link: http://lkml.kernel.org/r/20180122103947.GD2228@hirez.programming.kicks-ass.net
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Signed-off-by: Zheng Yejian <zhengyejian1@huawei.com>
---
 kernel/futex.c | 7 +++----
 1 file changed, 3 insertions(+), 4 deletions(-)

diff --git a/kernel/futex.c b/kernel/futex.c
index 199e63c5b612..70ad21bbb1d5 100644
--- a/kernel/futex.c
+++ b/kernel/futex.c
@@ -2248,10 +2248,6 @@ static int __fixup_pi_state_owner(u32 __user *uaddr, struct futex_q *q,
 
 	oldowner = pi_state->owner;
 
-	/* Owner died? */
-	if (!pi_state->owner)
-		newtid |= FUTEX_OWNER_DIED;
-
 	/*
 	 * We are here because either:
 	 *
@@ -2309,6 +2305,9 @@ retry:
 	}
 
 	newtid = task_pid_vnr(newowner) | FUTEX_WAITERS;
+	/* Owner died? */
+	if (!pi_state->owner)
+		newtid |= FUTEX_OWNER_DIED;
 
 	if (get_futex_value_locked(&uval, uaddr))
 		goto handle_fault;
-- 
2.25.4

Re: [PATCH 1/1] futex: Fix OWNER_DEAD fixup

[Greg KH]

On Mon, Feb 22, 2021 at 12:06:18PM +0800, Zheng Yejian wrote:
> From: Peter Zijlstra <peterz@infradead.org>
> 
> commit a97cb0e7b3f4c6297fd857055ae8e895f402f501 upstream.

Why is this not also needed in 4.9.y?  If so, please also provide a
backport there so I can apply this one too.

thanks,

greg k-h

Re: [PATCH 1/1] futex: Fix OWNER_DEAD fixup

[Zhengyejian (Zetta)]

On 2021/2/22 18:16, Greg KH wrote:
> On Mon, Feb 22, 2021 at 12:06:18PM +0800, Zheng Yejian wrote:
>> From: Peter Zijlstra <peterz@infradead.org>
>>
>> commit a97cb0e7b3f4c6297fd857055ae8e895f402f501 upstream.
> 
> Why is this not also needed in 4.9.y?  If so, please also provide a
> backport there so I can apply this one too.
> 

I've sent a backport for 4.9.y, please review it.
Thanks~

Re: [PATCH 4.4.257 0/1] Bugfix for ad4740ceccfb ("futex: Avoid violating the 10th rule of futex")

[Greg KH]

On Mon, Feb 22, 2021 at 12:06:17PM +0800, Zheng Yejian wrote:
> *** BLURB HERE ***

No blurb?  Why is this needed?

Re: [PATCH 4.4.257 0/1] Bugfix for ad4740ceccfb ("futex: Avoid violating the 10th rule of futex")

[Zhengyejian (Zetta)]

On 2021/2/22 14:40, Greg KH wrote:
> On Mon, Feb 22, 2021 at 12:06:17PM +0800, Zheng Yejian wrote:
>> *** BLURB HERE ***
> 
> No blurb?  Why is this needed?
> 
> .
> 

This patch may fix the following bug:

Link:
https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/tree/kernel/futex.c?h=linux-4.4.y&id=788437ba4c80d0d5e32ceaa28f872343e87236f5

     > static int __fixup_pi_state_owner(u32 __user *uaddr, struct 
futex_q *q,
     > 				  struct task_struct *argowner)
     > {
     > 	struct futex_pi_state *pi_state = q->pi_state;
     > 	struct task_struct *oldowner, *newowner;
     > 	u32 uval, curval, newval, newtid;
     > 	int err = 0;
     >
     > 	oldowner = pi_state->owner;
     >
     > 	/* Owner died? */
     > 	if (!pi_state->owner)
     > 		newtid |= FUTEX_OWNER_DIED;
Variable "newtid" is used without initialized.

Re: [PATCH 4.4.257 0/1] Bugfix for ad4740ceccfb ("futex: Avoid violating the 10th rule of futex")

[Lee Jones]

On Mon, 22 Feb 2021, Zheng Yejian wrote:

> *** BLURB HERE ***

You need to replace this place-holder with your own description of the
problem, how it's being solved and why this particular solution was
selected.

> Peter Zijlstra (1):
>   futex: Fix OWNER_DEAD fixup

To clarify, this *is* the right fix.  Thanks for taking the time.

>  kernel/futex.c | 7 +++----
>  1 file changed, 3 insertions(+), 4 deletions(-)

Reviewed-by: Lee Jones <lee.jones@linaro.org>

-- 
Lee Jones [李琼斯]
Senior Technical Lead - Developer Services
Linaro.org │ Open source software for Arm SoCs
Follow Linaro: Facebook | Twitter | Blog

Re: [PATCH 4.4.257 0/1] Bugfix for ad4740ceccfb ("futex: Avoid violating the 10th rule of futex")

[Greg KH]

On Mon, Feb 22, 2021 at 08:43:18AM +0000, Lee Jones wrote:
> On Mon, 22 Feb 2021, Zheng Yejian wrote:
> 
> > *** BLURB HERE ***
> 
> You need to replace this place-holder with your own description of the
> problem, how it's being solved and why this particular solution was
> selected.
> 
> > Peter Zijlstra (1):
> >   futex: Fix OWNER_DEAD fixup
> 
> To clarify, this *is* the right fix.  Thanks for taking the time.
> 
> >  kernel/futex.c | 7 +++----
> >  1 file changed, 3 insertions(+), 4 deletions(-)
> 
> Reviewed-by: Lee Jones <lee.jones@linaro.org>

Now queued up, thanks.

greg k-h