[tip:perf/core] x86: Avoid calling arch_trigger_all_cpu_backtrace() at the same time

From: tip-bot for Dongdong Deng
Date: Wed Jan 05 2011 - 09:08:18 EST

Commit-ID: 554ec063982752e9a569ab9189eeffa3d96731b2
Gitweb: http://git.kernel.org/tip/554ec063982752e9a569ab9189eeffa3d96731b2
Author: Dongdong Deng <dongdong.deng@xxxxxxxxxxxxx>
AuthorDate: Tue, 4 Jan 2011 22:38:08 -0500
Committer: Ingo Molnar <mingo@xxxxxxx>
CommitDate: Wed, 5 Jan 2011 14:22:57 +0100

x86: Avoid calling arch_trigger_all_cpu_backtrace() at the same time

The spin_lock_debug/rcu_cpu_stall detector uses
trigger_all_cpu_backtrace() to dump cpu backtrace.
Therefore it is possible that trigger_all_cpu_backtrace()
could be called at the same time on different CPUs, which
triggers and 'unknown reason NMI' warning. The following case
illustrates the problem:

set "backtrace_mask" to cpu mask
generate NMI interrupts generate NMI interrupts ...
\ | /
\ | /

The "backtrace_mask" will be cleaned by the first NMI interrupt
at nmi_watchdog_tick(), then the following NMI interrupts
generated by other cpus's arch_trigger_all_cpu_backtrace() will
be taken as unknown reason NMI interrupts.

This patch uses a test_and_set to avoid the problem, and stop
the arch_trigger_all_cpu_backtrace() from calling to avoid
dumping a double cpu backtrace info when there is already a
trigger_all_cpu_backtrace() in progress.

Signed-off-by: Dongdong Deng <dongdong.deng@xxxxxxxxxxxxx>
Reviewed-by: Bruce Ashfield <bruce.ashfield@xxxxxxxxxxxxx>
Cc: fweisbec@xxxxxxxxx
LKML-Reference: <1294198689-15447-2-git-send-email-dzickus@xxxxxxxxxx>
Signed-off-by: Ingo Molnar <mingo@xxxxxxx>
Signed-off-by: Don Zickus <dzickus@xxxxxxxxxx>
arch/x86/kernel/apic/hw_nmi.c | 13 +++++++++++++
1 files changed, 13 insertions(+), 0 deletions(-)

diff --git a/arch/x86/kernel/apic/hw_nmi.c b/arch/x86/kernel/apic/hw_nmi.c
index 2b40a60..72ec29e 100644
--- a/arch/x86/kernel/apic/hw_nmi.c
+++ b/arch/x86/kernel/apic/hw_nmi.c
@@ -28,10 +28,20 @@ u64 hw_nmi_get_sample_period(void)
/* For reliability, we're prepared to waste bits here. */
static DECLARE_BITMAP(backtrace_mask, NR_CPUS) __read_mostly;

+/* "in progress" flag of arch_trigger_all_cpu_backtrace */
+static unsigned long backtrace_flag;
void arch_trigger_all_cpu_backtrace(void)
int i;

+ if (test_and_set_bit(0, &backtrace_flag))
+ /*
+ * If there is already a trigger_all_cpu_backtrace() in progress
+ * (backtrace_flag == 1), don't output double cpu dump infos.
+ */
+ return;
cpumask_copy(to_cpumask(backtrace_mask), cpu_online_mask);

printk(KERN_INFO "sending NMI to all CPUs:\n");
@@ -43,6 +53,9 @@ void arch_trigger_all_cpu_backtrace(void)
+ clear_bit(0, &backtrace_flag);
+ smp_mb__after_clear_bit();

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