Date: Tue, 06 Apr 2010 02:50:49 -0700 (PDT)
Message-Id: <20100406.025049.267615796.davem@davemloft.net>
To: fweisbec@gmail.com
Cc: sparclinux@vger.kernel.org, linux-kernel@vger.kernel.org, mingo@elte.hu,
       acme@redhat.com, a.p.zijlstra@chello.nl, paulus@samba.org
Subject: Re: Random scheduler/unaligned accesses crashes with perf lock
 events on sparc 64
From: David Miller <davem@davemloft.net>
In-Reply-To: <20100405194055.GA5265@nowhere>
References: <20100405065701.GC5127@nowhere>
	<20100405.122233.188421941.davem@davemloft.net>
	<20100405194055.GA5265@nowhere>
Mime-Version: 1.0
Content-Type: Text/Plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 4973
Lines: 127

From: Frederic Weisbecker <fweisbec@gmail.com>
Date: Mon, 5 Apr 2010 21:40:58 +0200

> It happens without CONFIG_FUNCTION_TRACER as well (but it happens
> when the function tracer runs). And I hadn't your
> perf_arch_save_caller_regs() when I triggered this.

I figured out the problem, it's NMIs.  As soon as I disable all of the
NMI watchdog code, the problem goes away.

This is because some parts of the NMI interrupt handling path are not
marked with "notrace" and the various tracer code paths use
local_irq_disable() (either directly or indirectly) which doesn't work
with sparc64's NMI scheme.  These essentially turn NMIs back on in the
NMI handler before the NMI condition has been cleared, and thus we can
re-enter with another NMI interrupt.

We went through this for perf events, and we just made sure that
local_irq_{enable,disable}() never occurs in any of the code paths in
perf events that can be reached via the NMI interrupt handler.  (the
only one we had was sched_clock() and that was easily fixed)

So, the first mcount hit we get is for rcu_nmi_enter() via
nmi_enter().

I can see two ways to handle this:

1) Pepper 'notrace' markers onto rcu_nmi_enter(), rcu_nmi_exit()
   and whatever else I can see getting hit in the NMI interrupt
   handler code paths.

2) Add a hack to __raw_local_irq_save() that keeps it from writing
   anything to the interrupt level register if we have NMI's disabled.
   (this puts the cost on the entire kernel instead of just the NMI
   paths).

#1 seems to be the intent on other platforms, the majority of the NMI
code paths are protected with 'notrace' on x86, I bet nobody noticed
that nmi_enter() when CONFIG_NO_HZ && !CONFIG_TINY_RCU ends up calling
a function that does tracing.

The next one we'll hit is atomic_notifier_call_chain()  (amusingly
notify_die() is marked 'notrace' but the one thing it calls isn't)

For example, the following are the generic notrace annotations I
would need to get sparc64 ftrace functioning again. (Frederic I will
send you the full patch with the sparc specific bits under seperate
cover in so that you can test things...)

--------------------
kernel: Add notrace annotations to common routines invoked via NMI.

This includes the atomic notifier call chain as well as the RCU
specific NMI enter/exit handlers.

Signed-off-by: David S. Miller <davem@davemloft.net>

diff --git a/kernel/notifier.c b/kernel/notifier.c
index 2488ba7..ceae89a 100644
--- a/kernel/notifier.c
+++ b/kernel/notifier.c
@@ -71,9 +71,9 @@ static int notifier_chain_unregister(struct notifier_block **nl,
  *	@returns:	notifier_call_chain returns the value returned by the
  *			last notifier function called.
  */
-static int __kprobes notifier_call_chain(struct notifier_block **nl,
-					unsigned long val, void *v,
-					int nr_to_call,	int *nr_calls)
+static int notrace __kprobes notifier_call_chain(struct notifier_block **nl,
+						 unsigned long val, void *v,
+						 int nr_to_call,	int *nr_calls)
 {
 	int ret = NOTIFY_DONE;
 	struct notifier_block *nb, *next_nb;
@@ -172,9 +172,9 @@ EXPORT_SYMBOL_GPL(atomic_notifier_chain_unregister);
  *	Otherwise the return value is the return value
  *	of the last notifier function called.
  */
-int __kprobes __atomic_notifier_call_chain(struct atomic_notifier_head *nh,
-					unsigned long val, void *v,
-					int nr_to_call, int *nr_calls)
+int notrace __kprobes __atomic_notifier_call_chain(struct atomic_notifier_head *nh,
+						   unsigned long val, void *v,
+						   int nr_to_call, int *nr_calls)
 {
 	int ret;
 
@@ -185,8 +185,8 @@ int __kprobes __atomic_notifier_call_chain(struct atomic_notifier_head *nh,
 }
 EXPORT_SYMBOL_GPL(__atomic_notifier_call_chain);
 
-int __kprobes atomic_notifier_call_chain(struct atomic_notifier_head *nh,
-		unsigned long val, void *v)
+int notrace __kprobes atomic_notifier_call_chain(struct atomic_notifier_head *nh,
+						 unsigned long val, void *v)
 {
 	return __atomic_notifier_call_chain(nh, val, v, -1, NULL);
 }
diff --git a/kernel/rcutree.c b/kernel/rcutree.c
index 3ec8160..d1a44ab 100644
--- a/kernel/rcutree.c
+++ b/kernel/rcutree.c
@@ -286,7 +286,7 @@ void rcu_exit_nohz(void)
  * irq handler running, this updates rdtp->dynticks_nmi to let the
  * RCU grace-period handling know that the CPU is active.
  */
-void rcu_nmi_enter(void)
+void notrace rcu_nmi_enter(void)
 {
 	struct rcu_dynticks *rdtp = &__get_cpu_var(rcu_dynticks);
 
@@ -304,7 +304,7 @@ void rcu_nmi_enter(void)
  * irq handler running, this updates rdtp->dynticks_nmi to let the
  * RCU grace-period handling know that the CPU is no longer active.
  */
-void rcu_nmi_exit(void)
+void notrace rcu_nmi_exit(void)
 {
 	struct rcu_dynticks *rdtp = &__get_cpu_var(rcu_dynticks);
 

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/