* FAILED: patch "[PATCH] rcu: Allow for page faults in NMI handlers" failed to apply to 4.9-stable tree
@ 2017-10-02 11:44 gregkh
2017-10-02 19:33 ` Steven Rostedt
0 siblings, 1 reply; 5+ messages in thread
From: gregkh @ 2017-10-02 11:44 UTC (permalink / raw)
To: paulmck, rostedt; +Cc: stable
The patch below does not apply to the 4.9-stable tree.
If someone wants it applied there, or to any other stable or longterm
tree, then please email the backport, including the original git commit
id to <stable@vger.kernel.org>.
thanks,
greg k-h
------------------ original commit in Linus's tree ------------------
>From 28585a832602747cbfa88ad8934013177a3aae38 Mon Sep 17 00:00:00 2001
From: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
Date: Fri, 22 Sep 2017 14:10:22 -0700
Subject: [PATCH] rcu: Allow for page faults in NMI handlers
A number of architecture invoke rcu_irq_enter() on exception entry in
order to allow RCU read-side critical sections in the exception handler
when the exception is from an idle or nohz_full CPU. This works, at
least unless the exception happens in an NMI handler. In that case,
rcu_nmi_enter() would already have exited the extended quiescent state,
which would mean that rcu_irq_enter() would (incorrectly) cause RCU
to think that it is again in an extended quiescent state. This will
in turn result in lockdep splats in response to later RCU read-side
critical sections.
This commit therefore causes rcu_irq_enter() and rcu_irq_exit() to
take no action if there is an rcu_nmi_enter() in effect, thus avoiding
the unscheduled return to RCU quiescent state. This in turn should
make the kernel safe for on-demand RCU voyeurism.
Link: http://lkml.kernel.org/r/20170922211022.GA18084@linux.vnet.ibm.com
Cc: stable@vger.kernel.org
Fixes: 0be964be0 ("module: Sanitize RCU usage and locking")
Reported-by: Steven Rostedt <rostedt@goodmis.org>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c
index 51d4c3acf32d..63bee8e1b193 100644
--- a/kernel/rcu/tree.c
+++ b/kernel/rcu/tree.c
@@ -888,6 +888,11 @@ void rcu_irq_exit(void)
RCU_LOCKDEP_WARN(!irqs_disabled(), "rcu_irq_exit() invoked with irqs enabled!!!");
rdtp = this_cpu_ptr(&rcu_dynticks);
+
+ /* Page faults can happen in NMI handlers, so check... */
+ if (READ_ONCE(rdtp->dynticks_nmi_nesting))
+ return;
+
WARN_ON_ONCE(IS_ENABLED(CONFIG_RCU_EQS_DEBUG) &&
rdtp->dynticks_nesting < 1);
if (rdtp->dynticks_nesting <= 1) {
@@ -1020,6 +1025,11 @@ void rcu_irq_enter(void)
RCU_LOCKDEP_WARN(!irqs_disabled(), "rcu_irq_enter() invoked with irqs enabled!!!");
rdtp = this_cpu_ptr(&rcu_dynticks);
+
+ /* Page faults can happen in NMI handlers, so check... */
+ if (READ_ONCE(rdtp->dynticks_nmi_nesting))
+ return;
+
oldval = rdtp->dynticks_nesting;
rdtp->dynticks_nesting++;
WARN_ON_ONCE(IS_ENABLED(CONFIG_RCU_EQS_DEBUG) &&
^ permalink raw reply related [flat|nested] 5+ messages in thread
* Re: FAILED: patch "[PATCH] rcu: Allow for page faults in NMI handlers" failed to apply to 4.9-stable tree
2017-10-02 11:44 FAILED: patch "[PATCH] rcu: Allow for page faults in NMI handlers" failed to apply to 4.9-stable tree gregkh
@ 2017-10-02 19:33 ` Steven Rostedt
2017-10-02 19:38 ` Paul E. McKenney
0 siblings, 1 reply; 5+ messages in thread
From: Steven Rostedt @ 2017-10-02 19:33 UTC (permalink / raw)
To: gregkh; +Cc: paulmck, stable
Paul,
Can you backport this patch to 4.9 and 4.4 if it is required. The other
patches I posted (and also failed to apply, but should be backported)
depend on this patch, if it is an issue with those kernels.
-- Steve
On Mon, 02 Oct 2017 13:44:37 +0200
<gregkh@linuxfoundation.org> wrote:
> The patch below does not apply to the 4.9-stable tree.
> If someone wants it applied there, or to any other stable or longterm
> tree, then please email the backport, including the original git commit
> id to <stable@vger.kernel.org>.
>
> thanks,
>
> greg k-h
>
> ------------------ original commit in Linus's tree ------------------
>
> >From 28585a832602747cbfa88ad8934013177a3aae38 Mon Sep 17 00:00:00 2001
> From: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
> Date: Fri, 22 Sep 2017 14:10:22 -0700
> Subject: [PATCH] rcu: Allow for page faults in NMI handlers
>
> A number of architecture invoke rcu_irq_enter() on exception entry in
> order to allow RCU read-side critical sections in the exception handler
> when the exception is from an idle or nohz_full CPU. This works, at
> least unless the exception happens in an NMI handler. In that case,
> rcu_nmi_enter() would already have exited the extended quiescent state,
> which would mean that rcu_irq_enter() would (incorrectly) cause RCU
> to think that it is again in an extended quiescent state. This will
> in turn result in lockdep splats in response to later RCU read-side
> critical sections.
>
> This commit therefore causes rcu_irq_enter() and rcu_irq_exit() to
> take no action if there is an rcu_nmi_enter() in effect, thus avoiding
> the unscheduled return to RCU quiescent state. This in turn should
> make the kernel safe for on-demand RCU voyeurism.
>
> Link: http://lkml.kernel.org/r/20170922211022.GA18084@linux.vnet.ibm.com
>
> Cc: stable@vger.kernel.org
> Fixes: 0be964be0 ("module: Sanitize RCU usage and locking")
> Reported-by: Steven Rostedt <rostedt@goodmis.org>
> Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
> Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
>
> diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c
> index 51d4c3acf32d..63bee8e1b193 100644
> --- a/kernel/rcu/tree.c
> +++ b/kernel/rcu/tree.c
> @@ -888,6 +888,11 @@ void rcu_irq_exit(void)
>
> RCU_LOCKDEP_WARN(!irqs_disabled(), "rcu_irq_exit() invoked with irqs enabled!!!");
> rdtp = this_cpu_ptr(&rcu_dynticks);
> +
> + /* Page faults can happen in NMI handlers, so check... */
> + if (READ_ONCE(rdtp->dynticks_nmi_nesting))
> + return;
> +
> WARN_ON_ONCE(IS_ENABLED(CONFIG_RCU_EQS_DEBUG) &&
> rdtp->dynticks_nesting < 1);
> if (rdtp->dynticks_nesting <= 1) {
> @@ -1020,6 +1025,11 @@ void rcu_irq_enter(void)
>
> RCU_LOCKDEP_WARN(!irqs_disabled(), "rcu_irq_enter() invoked with irqs enabled!!!");
> rdtp = this_cpu_ptr(&rcu_dynticks);
> +
> + /* Page faults can happen in NMI handlers, so check... */
> + if (READ_ONCE(rdtp->dynticks_nmi_nesting))
> + return;
> +
> oldval = rdtp->dynticks_nesting;
> rdtp->dynticks_nesting++;
> WARN_ON_ONCE(IS_ENABLED(CONFIG_RCU_EQS_DEBUG) &&
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: FAILED: patch "[PATCH] rcu: Allow for page faults in NMI handlers" failed to apply to 4.9-stable tree
2017-10-02 19:33 ` Steven Rostedt
@ 2017-10-02 19:38 ` Paul E. McKenney
2017-10-02 19:47 ` Steven Rostedt
0 siblings, 1 reply; 5+ messages in thread
From: Paul E. McKenney @ 2017-10-02 19:38 UTC (permalink / raw)
To: Steven Rostedt; +Cc: gregkh, stable
On Mon, Oct 02, 2017 at 03:33:17PM -0400, Steven Rostedt wrote:
>
> Paul,
>
> Can you backport this patch to 4.9 and 4.4 if it is required. The other
> patches I posted (and also failed to apply, but should be backported)
> depend on this patch, if it is an issue with those kernels.
No problem!
How do we determine whether or not it is required?
Do we need to ask one of Bruce or Nick? ;-)
Thanx, Paul
> -- Steve
>
>
> On Mon, 02 Oct 2017 13:44:37 +0200
> <gregkh@linuxfoundation.org> wrote:
>
> > The patch below does not apply to the 4.9-stable tree.
> > If someone wants it applied there, or to any other stable or longterm
> > tree, then please email the backport, including the original git commit
> > id to <stable@vger.kernel.org>.
> >
> > thanks,
> >
> > greg k-h
> >
> > ------------------ original commit in Linus's tree ------------------
> >
> > >From 28585a832602747cbfa88ad8934013177a3aae38 Mon Sep 17 00:00:00 2001
> > From: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
> > Date: Fri, 22 Sep 2017 14:10:22 -0700
> > Subject: [PATCH] rcu: Allow for page faults in NMI handlers
> >
> > A number of architecture invoke rcu_irq_enter() on exception entry in
> > order to allow RCU read-side critical sections in the exception handler
> > when the exception is from an idle or nohz_full CPU. This works, at
> > least unless the exception happens in an NMI handler. In that case,
> > rcu_nmi_enter() would already have exited the extended quiescent state,
> > which would mean that rcu_irq_enter() would (incorrectly) cause RCU
> > to think that it is again in an extended quiescent state. This will
> > in turn result in lockdep splats in response to later RCU read-side
> > critical sections.
> >
> > This commit therefore causes rcu_irq_enter() and rcu_irq_exit() to
> > take no action if there is an rcu_nmi_enter() in effect, thus avoiding
> > the unscheduled return to RCU quiescent state. This in turn should
> > make the kernel safe for on-demand RCU voyeurism.
> >
> > Link: http://lkml.kernel.org/r/20170922211022.GA18084@linux.vnet.ibm.com
> >
> > Cc: stable@vger.kernel.org
> > Fixes: 0be964be0 ("module: Sanitize RCU usage and locking")
> > Reported-by: Steven Rostedt <rostedt@goodmis.org>
> > Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
> > Signed-off-by: Steven Rostedt (VMware) <rostedt@goodmis.org>
> >
> > diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c
> > index 51d4c3acf32d..63bee8e1b193 100644
> > --- a/kernel/rcu/tree.c
> > +++ b/kernel/rcu/tree.c
> > @@ -888,6 +888,11 @@ void rcu_irq_exit(void)
> >
> > RCU_LOCKDEP_WARN(!irqs_disabled(), "rcu_irq_exit() invoked with irqs enabled!!!");
> > rdtp = this_cpu_ptr(&rcu_dynticks);
> > +
> > + /* Page faults can happen in NMI handlers, so check... */
> > + if (READ_ONCE(rdtp->dynticks_nmi_nesting))
> > + return;
> > +
> > WARN_ON_ONCE(IS_ENABLED(CONFIG_RCU_EQS_DEBUG) &&
> > rdtp->dynticks_nesting < 1);
> > if (rdtp->dynticks_nesting <= 1) {
> > @@ -1020,6 +1025,11 @@ void rcu_irq_enter(void)
> >
> > RCU_LOCKDEP_WARN(!irqs_disabled(), "rcu_irq_enter() invoked with irqs enabled!!!");
> > rdtp = this_cpu_ptr(&rcu_dynticks);
> > +
> > + /* Page faults can happen in NMI handlers, so check... */
> > + if (READ_ONCE(rdtp->dynticks_nmi_nesting))
> > + return;
> > +
> > oldval = rdtp->dynticks_nesting;
> > rdtp->dynticks_nesting++;
> > WARN_ON_ONCE(IS_ENABLED(CONFIG_RCU_EQS_DEBUG) &&
>
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: FAILED: patch "[PATCH] rcu: Allow for page faults in NMI handlers" failed to apply to 4.9-stable tree
2017-10-02 19:38 ` Paul E. McKenney
@ 2017-10-02 19:47 ` Steven Rostedt
2017-10-12 23:14 ` Paul E. McKenney
0 siblings, 1 reply; 5+ messages in thread
From: Steven Rostedt @ 2017-10-02 19:47 UTC (permalink / raw)
To: Paul E. McKenney; +Cc: gregkh, stable
On Mon, 2 Oct 2017 12:38:57 -0700
"Paul E. McKenney" <paulmck@linux.vnet.ibm.com> wrote:
> On Mon, Oct 02, 2017 at 03:33:17PM -0400, Steven Rostedt wrote:
> >
> > Paul,
> >
> > Can you backport this patch to 4.9 and 4.4 if it is required. The other
> > patches I posted (and also failed to apply, but should be backported)
> > depend on this patch, if it is an issue with those kernels.
>
> No problem!
>
> How do we determine whether or not it is required?
Can rcu_irq_enter() be called within a rcu_nmi_enter() in these
kernels? If not, that needs to be fixed. Because nmi's can still page
fault.
>
> Do we need to ask one of Bruce or Nick? ;-)
>
The bug I fixed is actually there, but lockdep doesn't complain. The
module code that uses synchronize_sched() which requires RCU to be
watching when preemption is disabled. RCU isn't watching is some of
these cases that are called. Recent kernels added rcu annotations into
these preempt disabled paths which caused lockdep to report the issue.
The older kernels that had the bug just didn't have lockdep complain
about it.
-- Steve
^ permalink raw reply [flat|nested] 5+ messages in thread
* Re: FAILED: patch "[PATCH] rcu: Allow for page faults in NMI handlers" failed to apply to 4.9-stable tree
2017-10-02 19:47 ` Steven Rostedt
@ 2017-10-12 23:14 ` Paul E. McKenney
0 siblings, 0 replies; 5+ messages in thread
From: Paul E. McKenney @ 2017-10-12 23:14 UTC (permalink / raw)
To: Steven Rostedt; +Cc: gregkh, stable
On Mon, Oct 02, 2017 at 03:47:48PM -0400, Steven Rostedt wrote:
> On Mon, 2 Oct 2017 12:38:57 -0700
> "Paul E. McKenney" <paulmck@linux.vnet.ibm.com> wrote:
>
> > On Mon, Oct 02, 2017 at 03:33:17PM -0400, Steven Rostedt wrote:
> > >
> > > Paul,
> > >
> > > Can you backport this patch to 4.9 and 4.4 if it is required. The other
> > > patches I posted (and also failed to apply, but should be backported)
> > > depend on this patch, if it is an issue with those kernels.
> >
> > No problem!
> >
> > How do we determine whether or not it is required?
>
> Can rcu_irq_enter() be called within a rcu_nmi_enter() in these
> kernels? If not, that needs to be fixed. Because nmi's can still page
> fault.
>
> >
> > Do we need to ask one of Bruce or Nick? ;-)
> >
>
> The bug I fixed is actually there, but lockdep doesn't complain. The
> module code that uses synchronize_sched() which requires RCU to be
> watching when preemption is disabled. RCU isn't watching is some of
> these cases that are called. Recent kernels added rcu annotations into
> these preempt disabled paths which caused lockdep to report the issue.
> The older kernels that had the bug just didn't have lockdep complain
> about it.
It turned out that one backport is needed for 4.2-4.4, and another
for 4.5-4.12. I sent both.
Thanx, Paul
^ permalink raw reply [flat|nested] 5+ messages in thread
end of thread, other threads:[~2017-10-12 23:14 UTC | newest]
Thread overview: 5+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2017-10-02 11:44 FAILED: patch "[PATCH] rcu: Allow for page faults in NMI handlers" failed to apply to 4.9-stable tree gregkh
2017-10-02 19:33 ` Steven Rostedt
2017-10-02 19:38 ` Paul E. McKenney
2017-10-02 19:47 ` Steven Rostedt
2017-10-12 23:14 ` Paul E. McKenney
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).