ARM: 7803/1: Fix deadlock scenario with smp_send_stop()
authorStephen Boyd <sboyd@codeaurora.org>
Tue, 30 Jul 2013 22:09:46 +0000 (23:09 +0100)
committer黄涛 <huangtao@rock-chips.com>
Mon, 5 May 2014 08:17:08 +0000 (16:17 +0800)
If one process calls sys_reboot and that process then stops other
CPUs while those CPUs are within a spin_lock() region we can
potentially encounter a deadlock scenario like below.

CPU 0                   CPU 1
-----                   -----
                        spin_lock(my_lock)
smp_send_stop()
 <send IPI>             handle_IPI()
                         disable_preemption/irqs
                          while(1);
 <PREEMPT>
spin_lock(my_lock) <--- Waits forever

We shouldn't attempt to run any other tasks after we send a stop
IPI to a CPU so disable preemption so that this task runs to
completion. We use local_irq_disable() here for cross-arch
consistency with x86.

Reported-by: Sundarajan Srinivasan <sundaraj@codeaurora.com>
Signed-off-by: Stephen Boyd <sboyd@codeaurora.org>
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
arch/arm/kernel/process.c

index 978002e5b406224ea52c5f80980d32ba5d010d3a..d5768a7516f2d72bfcef9566fed9a844a2fe070f 100644 (file)
@@ -255,6 +255,7 @@ void machine_shutdown(void)
  */
 void machine_halt(void)
 {
+       local_irq_disable();
        smp_send_stop();
 
        local_irq_disable();
@@ -269,6 +270,7 @@ void machine_halt(void)
  */
 void machine_power_off(void)
 {
+       local_irq_disable();
        smp_send_stop();
 
        if (pm_power_off)
@@ -288,6 +290,7 @@ void machine_power_off(void)
  */
 void machine_restart(char *cmd)
 {
+       local_irq_disable();
        smp_send_stop();
 
        /* Flush the console to make sure all the relevant messages make it